Technical Memorandum
R1 1 roduk(d by
,o
'
THIS DCUMENT iS BEST QUALI Y AVAILABLE. T"I COPY FURN1SHED TO DTIC CONTAINED
A SIGNIFICANT E ICH NUMBER DO OF NOT
t'CLItftt
0  l;,NlA
'I Nt IF
v ,IT IT,
'C't,0[td
If. I Mot,)
!0,
P I POP I
S_ C J[NlIT
("L A* S
I I.C,
01
The Johns Hopkins University Applied Physics Lab. 8621 Georgia Ave. 20910 Silver Spring, Md. The Motion of Ballistic Missiles
A
[It ,C r4lI I
Unclassified
NA
IVE! NOT F 5
7'p
LI T "0 r4 1S
( Fi't
rIFAIo',
"Irddfle
fllNI
1d . IMN. 1II*11"l')
L. S. Glover and J. C. 6
Ha.
EPrORT 0DTE
IHagen
7I. TOTAL NO. OF PAGES Tb, N0 O IEPS R
July 1971
CONT4ACT COR GRANT NO.
390
90, ORIGINATOR'S 4EPIORT N MIBER(SI)
20
N0001762C0604 P o EC TN0.
G 1164
Wb. 0 TH ER HEPORT N OIB5 (Aniy a(her thin report)
d,
II DISTIRIBUTION STATEMENT
This handbook on the motion of ballistic missiles has been prepared primarily for use in training new personnel charged with the responsibility of evaluating the performance of reentry bodies. Aerodynamic fundamentals and characteristics of the atmosphere are reviewed briefly, followed by discussions of the effucts on ballistic missile motion of gravity, earth rotation, the atmosphere, and initial body rates. The anomalous motion resulting from various types of mass and/or aerodynamic asymmetries is discussed in detail. Simple solutions are given for the impact dispersion resulting from asymmetries and nonstandard meteorological characteristics at the impact location. Various trajectory simulations and flight test analyses are also discussed. Numerical examples illustrate the material.
OFF
UNCLASSIFIED
Security Claunification
14.
KEY WORDS
Impact dispersion
Missile trajectory Numerical solution
:1
tt
*1
'1
I,
Security Classification
Technical Memorandum
THE JOHNS HOPKINS UNIVERSITY *APPLIED PHYSICS LABORATORY 8621 Georgia Avenue, Silver Spring, Maryland 20910
Operaitnq under Contract WoCol 7 (3if.C..l6o4 with the Depdrinitirt of the. NA"'
......
ThEl JOHNS
HOP~i(NS UNIVENIiVI
ABSTRACT
This handbook on the motion of ballistic missiles has been prepared primarily for upe in training new per
Lii
JOWN1 IhN
HOPKI4
UNIolv"ItV
APPSLIE
PHYSICS LABORATORY
ABSTRA CT
missiles This handbook on the motion of ballistic new pertraining in has been prepared primarily for use of evaluating the sonnel charged with the responsibility fundamenperformance of reentry bodies. Aerodynamic reviewed are tals and characteristics of the atmosphrre on ballistic effects the of discussions briefly, followed by rotation, the atmosphere, missile motion of gravity, earth
inotion resulting and initial body rates, The anomalous aerodynamnic asymfrom various types of mass and/or solutions are is discussed in detail. Simple
metries resulting from asymmegiven for the impact dispersion characteristics at tries and nonstandard meteorological simulations and the impact location. Various trajectory Numerical exflight test analyses are also discussed. amples illustrate the material.
,Ny
(:1

iii
I
)*
UNIVERSITY
CONTENTS
List of Illustrations List of Tables 1 2 3 4 4.1 4. 2 4.3 4.4 4.5 4.6 5 6 7 8 Introduction Aerodynamics Atmosphere Equations of Motion Effects of Gravity Effects of Earth Rotation Effects of Atmosphere
. . .
. .
. .
vii
xi 1.
5
. . . . . .
. . .
, ,
Trajectory Simulation . Flight Dynamics Data Analysis Constants and Conversion Equations Index ,
. .
.
.
.
1 l9.
tWIql JONhlO
'40lPM1PW6i UNIVIUSIITY
ILLUSTIIA TIONS i. 21 22 23 24 25 26 27 31 32 33 41 42 43 44 45 46 47 48 Variation of Viscosity with Temperaturc Variation of Viscwqity with .36 Temperature Cone Axial Force Coefficient, Flow, Cold Wall, 8 = 100 Cone Axial Force Coefficient, Flow, liot Wall, 6 = 10* Laminar 37 Turbulent 38 39 40 41 65 66 67 . . 89 90 91 92 .. 93 94 125 12(6
35
Pitch Damping Coefficient for 10 Semiangle Cone Roll Damping Coefficient, Cold Wall, 6 = 10* Laminar Flow,
Roll Damping Coefficient, Turbulent Flow, Hot Wall, 6 = 10 Typical Atmospheric Temperature . Profile and Altitude Regions Standard Atmosphere Characteristics Effect of Water Vapor on Density Range versus y 0 and V Initial y versus V and Range R
Flight Time versus Range, Energy Trajectory Range Sensitivity to V Range Sensitivity to Y 0 Variation of II(K) with K Variation of I (K) with K
.
I'
vii
N,
IT4g jaid
MOR~NIN uNher.3TY
1 27
.
versus K
128
.155
412
155
413
414 415 416 417 418 419 420 421 422 423 424 425 426 427 51 52 53
227
228 229 230 231 2312 233 2600 261 262 263 264K
Variation of F(X, D) with X, ID 0. 2 Roll LockIn Criteria, Body A Roll LockIn Criteria, Body B Roll LockIn Criteria, Body C & Decay Characteristics Significant Events Required for g L pr and w
.
.259 . .
20'.20
versus Altitude
Conditions for Roll Resonance at h = 1) 000Feet Required for LockIn and Spin tdrough Zero Roll Rate .. Boundaries of~ and for LockIn at First Riesonance . . Variation of r with Initial Roll Rate mm. Limiting Values of KS. L. Limiting Values of A/ E Variations of E with h, Standard Atmosphere 1962
. . .
265
266
viii
~ILv6A
IPUIWO
MAPYLSK
ILLUSTRATIONS (cont'd) 54 Dispersion Resulting from Wind and Deviation in Density 55 56 57 58 59 510 71 72 73 74 75 76 77 78 79 710 711 71 2 Dispersion Rcsulting from Meteorological Characteristics at High Altitude Probability of Encountering Zero Roll Rate Maximum Altitude for Zero Roll Rate Maximum Dispersion Resulting from Se and r/d Dispersion versus 0 0 Dispersion Probability . , 312 31 3 314 315 316 317 318
Bit Presentation Scheme, DITAP Flight Records . .353 Timing Traces, DITAP Flight Records Bit Presentation Scheme, DITAR Flight Records Timing Traces, DITAR Flight Records . Sample Roll Rate History Sample Lateral Acceleration History Sample Longitudinal Acceleration .. History Sample Pitch Rate History Angle of Attack Work Sheet . . . 354 355 356 357 358 359 360 361 362 . 363 364
Aerodynamic Pitch Frequency (Weathercock Frequency) Work Sheet * Amplification Factor Work Sheet Rotation of Maneuver Plane Work Sheet . . .

ix
I
* 'M[ JOMNIl UNIV[IRIYI't H'OPKIN3N SlPSlf,N
TAB1,ES
Mass and Aerodynamic Characteristics 1062 Standard Atmosphere Constants for the 1962 Standard Atmosphere Reentry Characteristics Values of 13 Values of F
Values of 15 Values of F
iK
ix
Sxi
1.3
~;
1. INTRODUCTION
APPLIED
HYSICS LABOrATORY
The term ballistic missile as used herein describes an!j body in motion that is unpowered and uncontrolled. The motioa may occur in either the atmosphere or the exnatmosphere. However, most of the interesting motion phenomena occur wiithin the atmosohere, and most of the discussions concern reentrytype vehicles entering the atmosphere at speeds from 10 000 to 25 000 ft/sec.
'l l( p'iimar 'yv purpose_) oft thiL, rclp()rt is to pre sent
material suitable for' training new personnel charged with the responsibility of evaluating the performance of reentry bodies; therefore, the subject matter is treated as simply as possible while retaining enough detail to explain the missile behavior. Although the material is written primarily for personnel unfamiliar with the detai's of missile motion, it provides experienced personnel with readily available equations suitable for estimating orderofmagnitude effects of phenomena frequently encountered in flight test analysis, performance, and preliminary design work. A brief discussion ot the aerodynamic characteristics of missiles and the nature of the atmosphere is followed by a discussion of vehicle motion as affected by gravity, earth rotatior., the atmosphere, and initial body rates. Also included is a detailed discussion of the anomalois motion resulting from various types of mass and aerodynamic asymmetries. Simple solutions are given for the impact dispersion resulting from asymmetries and nonstandard meteorological characteristics at the impact location. Simulations required for various types of studies and flight test analysis complete the subjects discussed. Numerical examples illustrate the material presented. In some sections (particularly parts of Section 4) the material is new and is given in considerable detail, Since some readers may not be interested in details, a summary of the material covered and the major conclusions derived is given at the beginning of each section (each subsection for Section 4).
"~
~~~~~
..
.
An attempt was made to use symbols and units of measure currently used by specialists in the various fields covered in this report. As a result, a particular symbol mayhave different definitions in various sections. A complete list of symbols is given at the beginning of each section (each subsection for Section 4), and conversion factors for units of measure are given in Section 8.
4
___________________________________________________________I
~APPLIED PH4YSICS
2. AEAODYNAMICS
*jI Ii
Symbol A CA CAB CA F CA
I)efinition axial ro''rce axial force coefficient base drag coefficient friction drag coefficient pressure drag coefficient
P
c, g. C p C Cm
0
refers to center of gravity rolling moment damping coefficient pitching moment coefficient pitching moment coefficient at 0
Cm q C CN CN el
normal force coefficient 00 dC N/dWat refers to center of pressure total drag drag resulting from skin friction reference diameter Knudsen number (see Eq. (27)) lift or body length pounds feet pounds pounds feet
c. p. D DF d K L
Jt
Symbol
L(p) I M
Definition
rolling moment resulting from roll rate (roll damping moment) moment arm for asymmetric force Mach number = V/Ior
lbft lbft pounds pounds pounds lb/ft2 lb/ft 2 lb/ft 2 rad/sec rad/sec
2
pitching moment resulting from pitch rate (pitch damping moment) normal force
asy
normal force resulting from an asymmetry normal force for a sym.metrical vehicle ambient pressure cone surface pressure flap surface pressure roll rate pitch rate dynamic pressure, (y/2)PM (1 /2)pV2 gas constant = 1716 Reynolds number, oVL/14 = VL/h reference area static margin surface area excluding the base area absolute ambient air temperature reference temperature for viscosity vehicle velocity
01l
P P PF p q q 1R Re S S.M. w T TO V
lb /ft2 ft /sec ft 2 ft 2 OR OR
ft/sec
i
OR
"8
A
a.4)
UIVglvTl
Symbol X
Typical Units
x x
g.
distance from vehicle nose to the c. g. distance from vehicle nose to the c. p.
AX
Y (w t
c. p. c. g. distance normal to the body surface vehicle total angle of attack local angle of attack of specific heats

x
feet
feet radians radians
v yratio
8
(
M Ao
AO
"9
APPLIED PHYSICS
LABIORATORY
The aerodynamic forces of interest for reentry body studies are discussed, The significance of the aerodynamic coefficients is reviewed, and equations are presented for estimating the magnitude of the coefficients for a sharpnosed conical hody. Several types of asymmetric cones arc nlso con Hidn(rcd. Numerical examples are given for a 10 serniangle cone having a length of 10
feet.
The major conclusions are: 1. The following aerodynamic forces and moments are important in the analysis of motion of a typical reentry body: v Force perpendicular to tho body centerline (normal force) and resulting moment about the body center of gravity (pitching moment) Force parallel to the body centerline (axial force) Moments resulting by virtue of body pitch, yaw, or roll rotational rate (pitch, yaw, or roll damping moments) Forces and/oi moments resulting from body asymmetries.
* *
2. For uncontrolled vehicles, it is important that the normal force act through a point (center of pressure) aft of the center of gravity. In this case the vehicle is said to be statically stable. The ratio of the distance between the center of gravity and the center of pressure to some characteristic body dimension (for example, length) is called the stability nialgin and Is usually expressed as a percentage of the characteristic length. 3. The aerodynamic forces and moments are represented by dimensionless coefficients, thereby facilitating evaluation of the forces and moments over a wide variation in environmental parameters (velocity, altitude, etc. ),
Twx
4, In general, the coefficients are functions of vehicle angle of attack and the similarity factors, Mach number  an indication of the effect of compressibility of the air, Reynolds number  an indication of the nature of the boundary layer on the vehicle, and the Knudsen number  an indication of the size of the vehicle relative to the mean free path of the ambient air particles.
12
1
":..' ~~~~~~~~~~~~~~~~~......................'.
......
:"
UNIVR~P91?
2.1 Aerodynamic Forces A body moving through air experiences forces aerodynamic forces  resulting from the motion. The aerodynamic forces of interest for a reentry vehicle are: 1.
2.
A.
The axial force (for a symmetrical vehicle) acts along the body centerline as shown in the sketch for a cone.
C.
".
This force is produced primarily by relatively high pressure acting on the conical surface (pressure drag) and In addition, a force produced by air viscosity (skin friction drag) exists on the conical surface.
A normal force (normal to the body centerline)'"
II
13
,NIVINIITY
velocity vector, The angle between the velocity vector and the body centerline is called the angle of attack, . The normal force results from high pressure acting on the windward side of the body compared to low pressure acting on the leeward side. The integration of this pressure difference over the cone surface is the normal force. The pressure differential across the body alsc produces a mornent about the body center of gravity, Poor convenienctin analysis, this aerodynamic moment is computed as the product of the normal force and a moment arm (X is the longitudinal location at 1h6ch X1 . ), where X the fotal normal foice acts and is called the center of pressure and Xc. g. is the longitudinal location of the center' of gravity. Assume that a vehicle has its center of pressure forward of the center of gravity and that some disturbance causes an initial angle of attack. This angle of attack results in a normal force, Since the normal force is forward of the center of gravity, the aerodynamic moment produced is in a direction to increase the angle of attack, This process continues, and the missile eventually tumbles. Similar reasoning will show that, for a center of pressure located aft of the center of gravity, the aerodynamic moment will always tend to cancel a perturbation in &, Therefore, a vehicle having a center of pressure forward of the center of gravity is said to be statically unstable; a vehicle having a center of pressure aft of the center of gravity is said to be statically stable. The value of the ratio(Xc, p.  Xc. g,/L w AX/L is called the static margin. The margin must be positive for uncontrolled vehicles and for high performance vehicles is usually 3% to 5%. The use of the word "normal" to define the force In the plane of ;T and perpendicular to the body centerline is used here to be consistent with the usual terminology found in the literature. It is hoped that there will be no confusion with telemetry terminology where "normal" force and 'lateral" force are the two components of the In an attempt to avoid normal force as defined above. confusion, in some sections of this report we have used
14
TNS JO NOHOPKINS.rNIIvuuTY
Mg.
The local velocity distribution produces a local anrgle of attack, &I, whose magnitude is given by the equation:
(X

1
*
X)q
The local angle of attack results in a. local normal force which, at all points along the body, tends to resist the pitching motion. The Integral of the local forces (the damingforce) is tosmall to be of significance in ballisoet h apn tic missile aaye.Hwvr which is the integral of the local forces multiplied by their moment arms, is very important in some portions of the trajectory as will be shown later.
15 
INIiV2 1ItY MOPKIN&I THC JO~HN% APPLIED PHYSIC5 LABOnATORY MARI&NIA t$11V SIPRING
Tile example cited above i.s just one type of damping effect. Any force that exists by virtue of the pitch rate is a pitch damping force, and these forces do not always re.sist body pitch motion. The forces that do resist the pitch motion are said to be stable forces (or moments), and those that reinforee the pitch motion are said to be unstable forces (or moments). The only roll damping force on a symmetrical body of revolution is the force produced by the viscosity of the air, (skin friction). This force always tends to retard the roll rate. Ideally, it would be desirable to calculate the oerodynamic forces accurately by theoretical means. However, thio is nut always possible, and it is necessary to measure the forces in ground tests by blowing air past a fixed body (wind tunnel), or by propelling the body through a stationary instrumented region (ballistic range), or sometimes by using a combination of both techniques. Bconomically it is not possLble to obtain measurements on the actual vehicle for actual reentry conditions of ambient pressure, temperature, velocity, etc. To circumvent this difficulty, the aerodynamicist defines the forces nnd moments in terms of aerodynamic coefficie'ts which npply over a wide range of environmental conditions. The following coefficients are of interest in the study of renritry body motion:
A  CA S
N m CNqS M(q) _C (. iSd
(22)
(23)
(24)
V 1 p

16
?b41
JOH$1404SHOkpIN UN'VtN6ITV
where q at(1 /2)pV2 (,Yf2)pM 2 an(_ is called dynamic pressure. Tile bar over the q is deleted In somic sections of this report where there is no possible confusion between the symbols for pitch rate and dynaminic pressure. The reference diniensions S anrd d niay be iany convenient dimensions. For a cone, they nre usuanlly tnken as the ba sec ja0, Rca d h118 edin ( met. er, and thusu~ referen~ce dimen
rrom dimensional nnalys Is, it may he shown that for a given ;; thle "'rody"namic coefficients CA, CN (and
'
TieIch numbor.
dspeeds
velocity (V) to the velocity of a sound wave (VylT). "'here are four significant Which number re,,gimes. At subsonic (M ft0 to 0. I), the acrdynnai coefficients are weakly dependent upon M. Tho transonic Mach number regime extends from thle Ma~rch number at which a local Mach number over the body fi1rs~t becomes sonic (MI 1) to the Mach nuinbr, nt which.1 the flowfield over the hody is predominantly supersonic. This Mach number range speed results in the, appetirance of shock waves, Trhe strength and location of the wnveF are highly sensitive to changes in Mach number. Therefore, in the transonic speed range, thle aeurodynnmic coefficients are usually sensitive to change in M. In the supernonic speed range (M. P 1. 5 to (6), thle shock wave stteength Is dependent upon Mach number but the shock location is somewhat insensitive to changes In Mach number so that the aerodynamic coefficients are moderately dependent upon M. The effect of
17
speed aerodynamic heating 12 moderate in the supersonic aerodynamic the 6), > (M range. At hypersonic speeds and coefficients tend to be relatively insensitive to M aerodynamic heating is severe. The Reynolds number, pVL/M = VL/v, is indicaof the air, tive of the importance of the dynamic viscosity viscous of result a As M4 (or kinernntic viscosity t). in the vicinity forces, a region of low velocity is produced below) (discussed flow of the body. In fact, for continuum at the zero is vehicle the to the velocity of the air relative called is velocity degraded vehicle surface. The region of the to adjacent air of layer the boundary layer. Since the an or r, stress, shear a body "sticks" to the surface, element of friction drag, dD, results, tending to reduce the velocity of the vehicle,
I
LAMINAR
TURBULENT LAMINAR BOUNDARY LAYER THICK(NESS TURBULENT BOUNDARY LAYER ,' BODY SURFACE "THICKNINI
~,dS dr
TA (" dy
(26).,
*cosity
,
Therefore, the viscous drag is proportional to the dynamic wall. The visviscosity and the velocity gradient at the only temperature air of function a nearly is very

''
18
}i
APPLIED PHYSIC
LAMOHATORY
and is given by Sutherland's equation plotted in Figs. 21 and 22. The velocity gradient is a strong function of the Reynolds number. At very high Reynolds numbers, the flow in the boundary layer contains much eddy or vortical motion. As a result of mixing, the high energy air near the outer porLion of the flow is transported in toward the surface. Since the layer of air adjacent to the wall sticks to the surface, this type of boundary layer has a very large velocity gradient at the surface and the skin friction is relatively high. This type of boundary layer is called a turbulent boundary layer. At sufficiently low Reynolds numbers, the eddy motion within the boundary layer disappeari, and the velocity gradient at the wall and skin friction become much less than those for Lhu turbulent case. This type of boundary layer is called a laminar bounda,'y layer. At low Reynolds numbers the flow tends to be laminar; at high Reynolds numbers the flow tends to be turbulent. The Reynolds number at whinh transition from laminar to turbulent flow occurs varies with body shape and type of trajectory. A typical value of Re at which transition begin to occur on a slender conical body
is 107.
The surface heating rate is proportional to the skin frictioi and therefore is greater for turbulent flow than for laminar flow. In fact, the heating rate in turbulent flow may be nearly an order of magnitude greater than that for laminar flow. The Knudsen number is the ratio of the mean free path of the ambient air particles to a characteristic dimension of the reentry vehicle. This ratio may be expressed in terms of Re and M by the relation K = 1.49 R eM (27)
For the very low values of density that exist early in the reentry phase, the particles of air are spaced at distances that are large compared to, say, the body length
19 
or diameter.
likely to encounter another air particle. This region of flow is called the free molecular flow regime. At a somewhat lowtr altitude, the air particles have frequent collisions with other air particles in the vicinity of the reentry body. However, there are insufficient particles in the film of air adjacent to the body surface to cause it to stick to the surface. Therefore, although a boundary layer exists, the film of air adjacent to the wall "slides" along the body surface, This flow regime is called the slip flow regime, At still higher densities, the air particles at the surface stick to the surface, is called the continuum flow regime, This flow regime
The characteristics of the aerodynamic forces and surface heating depend upon the type of flow that exiots on the body. There are no sharp boundaries between the various types of flow, but approximate boundaries are as follows: Continuum to slip 1i0
2
Slip to transition
W 11 1
m
Transition to free molecular M R 10 e
10
The transition regime considered here is a transition in basic flow regimes which occurs at Knudsen numbers of approximately unity. This transition should not be confused with transition from laminar to turbulent flow in the boundary layer. The Mach number, boundary layer, and basic fluid flow regimes are shown for a typical reentry body in Fig. 421, page 260. Note that practically all the reentry ,
20
I * '
i I~I i.i,:lt *l r A
phenomena for which aerodynamic data are required occur in the continuum flow regime; only this regime will be considered in the remairnder of this report. Sometimes the lift nnd drrnf4 components of the for nnnlysis. The relntionship between the liiftdmtn g systclin of forces nnd the normal forceaxinl force system is shown in the sketch.
SN
A
a4
From this force diagram, the lift and drag coefficients may be obtained from the following equations: C
L,
cos
 C
sin a
(28) (2y)
Differentiating Eq. (28) with respect to c, and noting that the variation of CA with ey is small, the slope of the lift curve at a = 0 is given by the equation: CL =CN CA (210)
21
__
""%,. .. t'.
2. 2 G(enural Chnaincteristics It is beyond the scope of this r'eport to provide proce(duI'cr5lo'f estilmting the naerodynalmlic coefficuients for n body of arbhira'ry shape, Ilowever, a few re mareks on the
genernl behavioi' of tie coefficients nae given in this seetio)n, and :ppi' xini mte equa tions applicanle to n shn rp nosed CWne :i 1,fiVe.ii ill SCct ea 2. 3. Axial 'rcee. For most bodies of interest, the axial foorce coefficient is weakly dependent upon c for small values of ' v(10 to 2*), strongly dependent upon M, andt moderately dependent upon R.0 . For a sharp slender cone, the order of magnitude of CA is 0 1. Normal Porce. The normal force coefficient is strongly dependent upon rv and M and nearly independent of R1. For small values of fy, CN varies linearly with n so that CN = CNrY. For a sharp slender cone, CN a. 2 per radinn. Center of Pressure. In general, the center of pressure is strongly dependent upon o' and M and weakly dependent upon Re. However, for small vwlues of rY, Xcp P is independent of ry. For a sharp slender cone,
Xe p:, (2/3)L,
Pitch lDnnipng. The pitch damping coefficient is strongly dependent upon M, weakly dependent upon & for small values of ry, and nearly independent of Re. Fora a sharp slender cone, a typical value of Cmq is 1 per radian. The negative value of CnMq indicates a stable damping coefficient. 1oll Damping, The roll damping coeffici.nt for a symmetrical vehicle is very small, negative, hltiily dependent upon Re and M, and weakly dependent upon r for small values of . For I a cone, a typical vwlue of the roll clamping coefficient is 0. 003 per radian,
:1
22
I
tk " "'4
'L
r\
?HU[ JO94NS NOP4dINg) UHIVEMITYI
2. 3 Approximate Equations for a Sharp Cone For a major portion of the reentry trajectory, the Mach number is in the hypersonLc regime. The aerodynamic coefficients are nearly Independent of Mach number, and good approximations to the aerodynamic coefficients for sharp nosed cones are given by relatively simple equations. The skin friction drag coefficient, CAF, is very sensitive to the cone surface temperature, to the type of boundary layer, and to surface roughness. To provide some idea of the order of magnitude of the coefficient, CAF, without becoming involved in details, equations are given below which apply to laminar flow with a cold wall (typical of early reentry conditions) and to a turbulent flow with a hot wall (typical of late reentry conditions), A smooth wall was assumed. An estimate of C, is given in which it is assumed irected along the vector Is that the shear stress + pr where pr is the circumferential velocity of a point on the cone surface. The circumferential component of Ir is proportional to C1 The following equations are applicable to a cone of half angle 8:
CN
 2 cos N2
28
(211)
Xc.p.
2sec
8 L
(212)
=C
AP
+C 2
AB
+C
(213) (213
(214)
2 sin
1,
CAB
M.
(215)
23
CA
1. 9
tai 8
2+sin 2)Cosl
(216)
CA2Co 8 tan
0.___ (1 +
M _;n 6
cc)1.
r
CA
01
alay
(2r1
hc teaymtrcfrei
sroduce
diastru neavte
24
At a 0, the protuberance results in a downward asymmetric normal force, Nasy. This force causes the nose to pitch up, producing an angle of attack. The angle of attack produces a normal force, Ns. The vehicle continues to pitch up until the moment about the center of gravity is zero. The zero moment condition is called the trim condition, and the values N and re that exist at this condition are called the trim normal force and trim angle of attack, respectively, It is convenient to define another dimensionless aerodynamic coefficient called the pitching moment coefficient, Cm, such that the moment about the center of gravity is given by!
M=CM q Sd.
(220)
From the above sketch (using a noseup moment as positive): asy 1 (s c.p. c.g.
Ns i asy q Sd
Ns IS(IRS Nasy S
(223)
224)
The term, Ns/lS, is the CN for a symmetrical vehicle. The Nasy term is usually negligible compared to Ns/qS
25
and may be ignored in Eq. (224). However, the effect of this term on Cm cannot be ignored. Let
asylaC
qSd Then (X x
C
0
C
N
C. .
d
(225)
then Eq,
" C
p. d
m 0
If Cmo and X., p, are independent of 9, linearly with a.
(226)
then Cm varies
Within the limitations imposed by the assumptions listed above, typical Cm characteristics for a stable symmetrical and a stable asymmetrical vehicle are shown in the sketch, page 27, Note that C is the value of Cm at o 0, and TR is
The value of
C
m 0 x
i 0 Z27 (227)
~TR
(X
CN
26
TH
JIOHISi "O
CL
ASYMMETRICAL VEHICLE
, .
SYMMETRICAL VEHICLE
I'
TR
(L/d)CN S,'M.
o(228)
An equa
(229) con
27
iN
Thi
'I
a 5 P1
ley
qA
'
~
.1.
The pressure acting on the flop and cone surfaces may be estimated using Newtonian theory which, in the simplest form, states that (for zero angle of attack):
PF "2
q
A"1
2 sin 2 ( + c)
(230)
=2
sin 28
(231)
may be divided The force resulting from the asymmetry into normal and axial force components. For small values of 6 and (, the axial component Is small compared to the normal component. Neglecting the axial component, the total increment in normal force is:
NasySF(PF
(232) i
IS d
28
3) (2q
 238
 p.
'
Cm
0
20. For this asymmetry, Eq. (234) shows that 0.0027. Another type of asymmetry for which Cmo can be
readily evalUL:Lcd Is one in which the body centerline is the arc of a circ*le, This type of asymmetry, for example, is an approximation to a distortion resulting from lateral loads or heating effects. For this asymmetry the moment is noarly a pure couple and Cmo may be expressed in turms of the change in slope of the body centerline from nose to base.
007
mo
O 0.008 Le
(235)
where A is in degrees.
29
Ii
..
.H
fTl
JIOHNSIHOPKIPN UNIVERIIIv
SILVIN SPMI2N MANYt.AND
2. 15 Examples
A typical reentry body will be used in many examples In later sections of this report. To illustrate the material in the preceding subsections, the aerodynamic data required for subsequent examples will be computed. Th r typical body is a 10footlong cone with a base diameter of 3. 5 feet. These dimensions correspond to a cone semiangle of about 100. The mass characteristics are assumed to be as follows: Weight X 1000 pounds
= 6. 27 feet aft of the nose
At hypersonic speeds and for angles of attack close to zero, the aerodynamic coefficients are as follows: SC.
, Xc p.
2 3 se c 1oo0
10 6,87
feet
(212)
(211)
N
2 cos
It is observed from Eqs. (213) to (217) that CA is a function of M, Re, wall temperature, and type of boundary layer. The valves of CA for two combinations of wall temperature and boundary layer type are shown in Figs. 23 and 24. The laminar flow, cold wall case is typical of conditions that would exist early in the reentry flight, say above 100 000 feet; the turbulent flow, hot wall case is typical of conditions which would exist loter in
.I
I
THE JOHN@ HOPKINS UNIVR"4IY
flight, say below 100 000 feet. As will be shown later, most of the phenomena of Interest occur for values of He from about 106 to 108. For these values of it
CA ft 0.i 1
A
At M = 10 approximately (60% Ot' CA results from presrure drag (CAP), 30% from skin friction (CAF), and about 10% from base drag (CAn). The percentages vary somewhat with Re, M, and type of boundary layer, but the given values are typical of the relative importance of the three sources of axial force for lowdrag reentry bodies. For blunted bodies, the pressure drag may account for nearly 100% of the total drag. Values of Cmq (Eq. (218)) are shown as a function of X , /L i'n Fig. 25, and values of C,(Eq. (2.19)) are shown in 'Figs. 26 and 27. p In the examples included in subsequent sections we will use typical values of the aerodynamic coefficients. The values of the rmass and aerodynamic characteristics that will be used are shown in Table 21.
31
,i
31
Til,
JOHNN* "O"'B
aN
BIBLIOGRA PHIY
21
S. A.
Schaaf :tlid L. Talbot, tlndbook of SuperMechanics of Rarefied Gnses, NAVORDSYSCOM Report 1488 (Vol. 5), 1"',n!o,
22
M. Tobak an, \ 1, . Wel'rend, Stability Derivatives of Cones nt Supersonic Speeds, NACA TN T7768, Se.pte,.,L.,.t_' ,
_"_'_
_,_..,
=,
iI
SILVI opTp
IA
LU
JI
( I
u...
I0 2L
I II
rfu
*35
N141VCSIIY
'I
4

3,1 X103
SLUGFTSE
!iQ2
To
619 o R

SUTHERLAND'S FORMULA
4+
191
0I
oi
1Qf3200
12
l96.7/TO
198.71
400
R00
12DO
1(oo
2000
2400
2800
3600
4000
TEMPERATURE (CR)
FIg, 2.2
36
,I
I...
_ _
,i
YMI jhWId
NOl.w6
UWi'4aus' 'V
0.42
0.3
0.2
1
k. A
106
0104
Sig, 2.3
10.
37
',2
.......
.....
I..... ,. .
......
r.....
...
. i M" 10 20
... '7 *
< 0.1
_____n_____
10 5
10 7
10u
10t
Fig, 2.4
38
4
I...
 38
S,'.,________,_,_._,___.
TII
*1
.1.2
0.4
I 0.54
I 0.58 X cg./L
I 0.62
I 0.66 0.70
0.60
FIg, 2"5
39
S......
.
..
....
....
.. ",
',',s..,' , . . ,',';:..t"..
'.
. op'AIToVY IJHN$
A'
o *La,
_P
,,e',.A TO
.0,04
0.04 o.030
M  20
l .o.02 0
40'
0.0003
T'
"
0.002
.0,00`1
lo'
10
FIg. 27
100
41
i
S
" ' ., :.:. r ',
. .. ...
,, "r. , . h." "" ,' ""'.g , .: r...l m .I,, 1 .
Table 21
Mass and Aerodynamic Characterlistics Masi3 Characteristics L 10 1 feet ft 2 S 9, 65 d m 3. 5 feet X g, . ,27 feet 6 Weight (W)  1000 pounds Mass (m)  31. 1 slugs 2 30 slugft x (Iv) Inertia of RAI moment 300 slugft z Pitch moment of inertia (Iy I)
ly
a2
0. 99
Ix
.2
0.079
=
"
/4%
CA 0.104 Cm C, p AX
0q 14,
42
*;,.
I,
3. ATMOSPHERE
4
AA
[
!: [  43 
lHIr
JIONNSI
JO
'40pINSI
UNIVSNS,?YlT
"NIVIPSI.Y
Symbol a
Napierian base
c
Coriolis force horizontal force resulting from a pressutre gradient acceleration of gravity scale height, RT/g 0 geometric altitude geopotential altitude static pressure of dry or moist air partial pressure of dry air saturation pressure at T slb/ft partial pressure of water vapor gas constant for dry air radius of earth
g H h
h
P P p p R
CT
2 2
~2
R RAOB's
ROCOB3's
gas constant for water vapor weather measurements in which a balloon is used as instrumentation carrier weather measurements in which a rocket is used as instrumentation carrier sea level temperature of dry or moist air dew point temperature
S. L. T T
R
OR
45
Symbol t V
V W dx, dy, dh Z c
pounds feet
percent
v
0
mean free path kinematic viscosity density of dry air x P/RT density of moist air earth rotation rate
P
pr
Subscripts In o refers to a reference altitude or to altitudes at which weather measurements are obtained refers to sea level conditions
4'
The atnisphore from se, level to urhout 400 000 feet Plays '1n imlpoi't:uit vole ill the moition it'~rumiir V vehicles. In this section the generna mi tur1 , of ti n till sphere Is considered, and equations required for studies concerning vehicle motion Lire discusised. The vnria.blvs of most concern are atmospheric, density, temperature, wind, and moisture profiles (vtitrat ions with altitude), The 19~62 Standard Atiosphere i~s used in many anialyses. Some of the charactvristics of this atmosphere are prosented, and differences from othow types of sttandalrd days are noted. Tho major conciusionis iru1 . Most phenomena of interest in the analysis of reentry body motion occur in an altitudo region cnfled the homosphere (son level to approxiimately 300 000 feet), In this region the atmosphere obeys two laws of considerable importance. The perfect gas law Provides a r'elationship between pressure, temperature, and density; the perfect gas law in conjunction with the hydrostatic equilibriumnlaw provides a relationship between density, temperature, and altitude, The hydrostatic law 1is frequently used as an interpolation equation for tabulnr values of density versus altitude, permitting relatively few volues of the independent variable for a given aiccuracy. Trhis considera tion is of particular importance in trajectory simulation work. 2, Winds are obtained from direct measurements (b,,alloon observations, or PKA01Vs, for the lower aitmosphere and rocket observationsor lROCOl3's, for the uipper atnmosphere), from forecasts based on ohservations~or from l'orecasts based on statistical analyses of many previous observations (climatological data). Thiese sources are listed in the order of decreasing accuracy, and all sources 47
3. If the mo.."txirc content of the air is high, thre perfect gas law nust be applied using partial, pressures of air and water mixtures with appropriate gas constants. Sometimes moisture is accounted for by using an effective (or virtual) temperature. The effect of water vapor is to decrease the density when it is compax'ed with a computed density in which the moisture content is neglected. Neglect of the moisture effect results in a density error of a few percent at most, and the error decreases rapidly with decreasing altitude and decreasing temperature. For most analyses this effect may be ignored.
4BI
Ii
 48
~tH9
ie s; dlivided rinto thi'wee br)oad roltitude T1he ~ m~p ry ridnterplaneta bands cal led tile haro sphiere, uxc ~pher e, :11 uCC1LP to h (xisti froml S(e:ilvl g,18 (tPIg. 3i1). T110' brosp13)IM.'O eoper lih. about 500 km1 (1. (65 . 10 feet or 275 rim i).
Cxi~tS fV01m 500 to ;1h(oot (GO 000 kmI
e1x0Hphere'L,
jso
III the (331 000 Mi110. idlyI klow' tjiot. iton lii thle Molecules enltering
nt
baise of this region dlescribe ballistic trajectories, and some, tra vel beoynd the gin Vi taltLIonal 1'i(ld of theL planet, III thilik region, temlper~ature ins rico significnr ice since there is rio Mnxwollalan distribution of vel'ocity. Int er planetary gas ucists at tittitudos above about GO0 000 km1.I The barosphere is subdivided into the troposphere, stratosphere, niesosipherc, and thermosphere with thle tropopnuse, it~ratopnuse, mresopause, and thlermlopause locnted between theose regions, Lis shown in FPig. 31, These altitude bands are, baised upofl thle temperature variation with altItude,. Tile region from sea lev~el to the top of tile mesophere is also called the homosphere. The behavior of each altitude band varios somewhat with cartint latitude, Many types of standard atmospheres are in commion use; only three will be considered in this section. Thle 1062 Standard Atmosphere is typical of middle latitude locations; the Tropical Atmosphere is typical of low latitude locations (01 to 20' N. latitude), and the Polar Atmosphere is typical of high latitude locations ~(north of 60' N. latit~ude). In thle troposphere the variation of temperature with altitude is strongly influenced by the earthi acting as a heat source. The temperature, in general, decreases with increasing altitude. The top of thle troposphere occurs at an altitude of 36 000 feet for the Standard Atmosphere, at 55 000 feet for tile Tr'opical Atmosphere, anid at 83 000 feet for the Polar Atmosphere. Essentially all weather phenomena occur in the troposphiere.
* *
The.( tropopause is the top of' the troposphere Ind is an altitude of minimiumn temperature. i'or the Standu rd Atmosphere the temperatlure is constant from 3(3 000 to 66i 000 feet, where it begins to rise; for the tropicail ntmnosphere the temprnlnture begins to rise rit 55 000 feet; for the polar' atmiosphecre the temiperatureVL is (mnstant to Wi 000 feet, Where it b. girls to rise. The teniperatLre rise is a
result of ;I)5Iovploiili ot iltI8\'i(1i'ht sulni, rtidiations by
ozone. The temper'ature Conitinlues to riisc with mecca sing altitude to about 1.563 000 feet for the Standard Atmosphere. Characteristics to suifficiently high a.ltitudes re not avaU.aible for the tropical and polar, atinosphereSg. Above the stratosphere the teiperature again decreases with increasing altitude., reaching ;i mninimum temperature at the inesopause. Pox' the Standard Atmosphere, this 1ltitud.a is 263 09J0 feet. Above the mesopause the tempera
therniopause, which occurs at approximately 1 650 000 feet (275 nmi). Thu temperature rise is caused by the absorption of ultravio~let rays with a wavelength of 100k. The homosphere (sea level to 300 000 feet) is the altitude band of primary interest for analysis of reontry
body motion.
characterized by constant proportions of N 2 , 021 and argon, and the atmosphere obeys the hydrostatic equilibriumn and perfect gas equations. For analysis of reentry vehicles, tile atmospheric proJerty of most interest is air density; of secondary interest is air temperature. The ratios of density f'w the tropical arid polar atmospheres to the density for the 10)62 Standard Atmo)sphere are shown in Fig. 32. It may be expected that the tropical. atmosphere would have a lower density and the polir atmosphere a higher den
'
I
'50
over, this is true only for altitudes b~elow about 20 000 to 30 000 feet. Therefore, depending upon thQ vehicle design and trajectory, the tropical day may be equivalent
Ho0wI
LRW.....
Starviard Dlay, and the polar day may be, equivalent to An atmosphere of lessor density than thnt of the 1062 Standard Da y. The toemperature, in addition to the effect on density, qffects Cie sp.)cd of' sound, and thas Mach number VI/a). rhe spceed of sounfd ( =iWT) is1 proportional to the s qua rc VMut of' tiw nvnibi eat ternipe 1'tu re aad tli o'e c haraneteristics are aLso shown in Fiig. 32, Notu that the sp,ud of sound at a given altitude does not vary from. one standard atmosphure to anothev by more than about 7. 5l%. This amount of variation is r'elati~vely unimportant. How* ever, extreme variations in temperature from the I J62 Standard Atmosphere, such as those of the winter months it northern latitudes, may have a significant effect on a reentry body trajectory by affecting the Mach number history. Some, characteristics of the 1962 Standard Atmosphere are given as a function of altitude in Table 31. More elaborate tables Lire given In R~ef. 31. Similar data for the tropical and polar days are given in Reis. 3 and 3,3. 3. 2 Variation of Density and Temperaturc with Altitude Consider an elementary volumne of air (page .52) of base area dxdy and height dh, If the elemnent: of volumo is in static equilibrium (the acceleration of the elument of volume is zero), then the following equation applies: dxdy(P + dp dh) Pdxdy + W The weight of the volume of air is given by:
:
W = gdxdydh
Therefore, dP

gdh
(31)
li1t
SPRING
MHATLAND
p+dP
R
dh
dx
This equation Is called the hydrostatic equilibrium equation, and all three standard atmospheres dlicussed in this section obey this equation, A second equation that is valid in the homosphere is the perfect gas law which states!
P = RT , (32)
(33)
Substituting Eq. (33) into Eq. (31) and expressing g in terms of go: p T RTg
By definition:
dh G go0 dh (35)
52
......................................................................
where hG is called the geopotential Rltitude and h is the geometric altitude. If g is assumed to vary inversely with the square of the distance from the center of the earth: R
0
('6)
I
,
+hh
(3"7 )
or
h h G
1 "
hG
(38)
In the remainder of this section, the solution to Eq. (34) will be given in terms of the geopotential altitude. However, the results may be obtained in terms of h by using Eq. (38). It should be noted that, since Ro f 2. 09 x 107 feet, h f hG for low altitudes. Even at h 1 100 000 feet, hG and h differ by only 477 feet. In terms of hG, Eq. (34) becomes: dT = d_ + p T go  dh RT G (39)
'4
The closed form solution to Eq. (39) depends upon the nature of the variation of T with hG. For practical applications it is necessary to consider only two cases, a constant temperature (isothermal) relation and a linear relation.
 53 
hic APPLI rJ
JOHNi
Consider an altitude interval from hGn to hG. grating over this altitude band, Eq. (39) becomes:
Inte
go n n
hG Idh hG n
(310)
If T
is constant,
h
P "P exp
Gn
h
(311 131,l)j
where RT' H  90 If T varies linearly over the altitude band, then: +dT T nT+(312)
d h
(h
 h
Gn
(313)
.2
For calculations, it is frequently desirable to define the density and temperature variations with hG by an equation rather than to use tables, If the atmosphere obeys the hydrostatic equation, it may be divided into altitude bands for which the temperature is accurately defined by linear variations of T with hG. Then the density and temperature within a given band are defined using Eqs. (311), (313), and (314) and the conditions at either the lower or upper boundary of the band.
54 
A!
,
..
 ':
. ,. .
.. l'
. ,.
Iy
The characteristics of the 1962 Standard Atmosphere may be defined to within 0. 5% accuracy using Eqs. (311), (313), and (314) and the data in Table 3.2. For simple calculations, it is frequently assumed that the entire atmosphere is at a constant tempernture. Such an atmosphere is called an isothermal or exponential atmosphere. In this case, Eq. (311) becomes
Values of
and
racy for pin the altitude region of most interest for the particular study. The quantity H is called the scale height. From Eq. (311), the scale height is the change in altitude corresponding to a change in density (or pressure, since T is constant) by a factor of 1/e. 3.3 Nonstandard Atmosphere
Frequently it is necessary to consider atmospheres other than the Standard Atmospheres. The quantities of primary interest for studies of reentry body motion (neglecting such specialized studies as the effects of wind shear and wind gust) are density, temperature, horizontal wind direction, and horizontal wind speed. These data are available from observations, forecasts, or' climatological characteristics. The latter data are characteristics based on statistical analyses of many observations. Observations are made in two altitude regions. Balloons are used to obtain temperature, pressure, dew point, and wind data at altitudes from the surface up to the highest altitude possible. Frequently, the upper limit in altitude is about 30 000 feet. Sometimes the data extend to observations at about 100 000 feet. These observations are called RAOB's. In the northern hemisphere, RAOB's are obtained daily at OOZ and I 2Z hours at many observation points. Rockets are used to make observations of
55 
 1
"
U
temperature, density (sometimes), and winds as a function of geometric altituie for altitudes from about 60 00nf to 200 000 feet. Data from these soundings are called ROCOB's and are made only at a very few locations, usually near 12Z hours Mondays, Wednesdaye, and Fridays. 3. 3. 1 RAOB )atDryAir The RAOB pressure and temperature datta are converted to density data by Eq. (32), The geopotential altitude corresporiding to the observed P and T are calculated using the hydrostatic equation. Since data aro obtained at a large number of altitudes, the temperatures from one measurement to the next may be averaged and used as a constant so that Eq. (3i1) applies and may be written in the form: Tn+I) h Gn+l uhG hn + R(Tn 2 + gn+n n (316)
where n corresponds to the measurement at one altitude and n+1 corresponds to the measurement at tne succeeding altitude, Equation (316) is evaluated in stepwise fashion starting at the surface conditions where hGn is the surface elevation of the observation location. 3. 3. 2 RAOB Data  Moist Air The values of density computed uaing Eq. (32) are valid provided the effect of water vapor may be neglected. The effect of water vapor on density may be evaluated by writing the equation for the density of the airwater mixture as the sum of the partial densities: PA + "w RT R T
w
(317)
I I,
a.
,1
IN
i
where P "
A + Pw
0. 622:
Pw 0( 1

o 378
.(319) ,)
If the moisture content is given in terms of relative humidity, 0  Pw/Ps, rather than Pw, then,
,'
1  0. 378
(320)
The water vapor pressure, Pwo is the saturation pressure at TD, and P 5 is the saturation pressure at T. Empirical
equations for the saturation vapor pressures are:
P.in. exp 2. 65
Ps i
exp
5T
6i5
whc2.
gne
rThe effect of water vapor is to decrease the dens~ity compared to a computed density which ignores the effect of moisture. given by: The percent difference in density is
37,.
37
o D
exp 22. 65  99
(322)
Tx
A plot of the error as a function of altitude anid dew point is given in Wig. 33. Also shown are the standard
57

temperatures for the 1962 Standard Day and the Standard Tropical Day. For these standard days, the effect of 100% relative humidity (TD = T) on percent change in density is maximum at sea level and is about 0. 7% for the 1962 Standard Day and about 1. 8% for the Standard Tropical Day. Only for very hot humid days is water vapor likely to have any significant effect on density, and even this effect is significant only for very low altitudes.
0n+1
n [R(Tn
323)
where p+l is obtained in stepwise fashion using a known initial value of density (from RAOB data, for example), measured T, and hG computed from Eq. (37) using measured 3.3.4 h. Winds
A wind is defined in magnitude (speed) and direction from which it is blowing, measured clockwise from the north. In some current forecast procedures, the predicted winds are not based on measured winds but upon measured pressures and temperature, using the geostrophic wind equations, The geostrophic wind is obtained by balancing the horizontal pressure force acting on an elementary mass
 58
of air by the Coriolis force, The Coriolis force is perpendicular to the horizontal velocity of the elementary mass and, in the Northern Hemisphere, is directed to the right (to the left in the Southern Hemisphere). The magnitude of the force acting on an elementary mass of air, pdxdydh, is given by the equation:
Fc
2V wto
sin 0 dxdydh ,
(324)
A horizontal pressure force acts on the elementary mass whenever a horizontal pressure gradient exists. Consider the following horizontal high pressure system. Let x be the horizontal distance perpendicular to the isobars, y the horizontal distance parallel to the isobars, and h the vertical dimension,
Vw
ISOBAR
Fi
FC
X
HIGH
PRESSURE
dx
The pressure on the two surfaces of area dxdh are equal, so that the net pressure force acting on the mass in a direction parallel to the isobar is zero. The pressure on the two dxdy surfaces are unequal, but the weight of air is just balanced by the pressure force (hydrostatic
59
?He JOWNG
HOPKINS
UNi~aRSITV
equilibrium is assumed), so that no net vertical force exists. Let the pressure acting on the dydh surface that L~s farthest from the high pressure center be P. Then the pressure oui the dydh surface nearest the center is Fr + aF dx. The corresponding forces are Pdydh and (P + dx)dydh. F u~ Therefore, the net force is,
dxdydh,(2)
Isphere),
The total horizontal pressure force is perpendicular to the isobar and directed away from the high pressure, Since the geostrophic wind is obtained by a balance of the pressure and Coriolis forces, the Coriolis force is perpendicular to the isolbar and acts toward the high pressure center, am ehown in the sketch, For a Coriolis force In this direction, the wind maust be directed along the isobar with high pressure to the right (for the Northerm Hemi
Equation (326), the geostrophic wind equation, Is a,curate except for the first few thousand feet above the surface where friction effects cause the wind to be inclined away from the Isobar toward low pressure and the magnitude is lower than that given by Eq. (326), Typically, the mnaximum deviation in direction is 250, and the maximum devia. tIon in speed is a factor of onethird to one*half the geostrophic wind, depending upon surface roughness. In addition, the equation is not valid at latitudes near the equator, The equation is valid for the Southern Hemisphere, but the direction of the velocity along the isobar is opposite to that discussed for the Northern Hemisphere,N Equation (326) is applied at various altitude levels to obtair. the variation of wind speed with altitude. The
f.
~~6
IfIlbulll
wind direction is obtained from the slope of the isobar at the location of interest. Therefore, accuracy in wind characteristics ,'equires accurate estimates of density, horizontal pressure gradient, and shape of the isobars.
tV
.i.
51.1
,'
1!
T4e JOMNS 400IWO uI1N'1I0I6TV
REFERENCES
"31
U. S, Standard Atmosphere,
sponsorshlp of National Aeronautics and Spoce Adminiistration, United States Air Force, and
United States Weather Bureau, Washington, D. C., December 1962, 32 33

Mil.
PI
I
K
1'
__ _________ . 4 .,.%i1 
;I
T1,9 JOHN% HOPKINs UNIVtP5Ily
INTERPLANErARY
GAS
000
NOTE: NOT TO SCALE
FEET33
EXOSPHIFR,
000 rirnr
l
320

'r~HE=RMVOP'AUSE
tl
275 nrmi
i
T"HEAMOSPHEE /MESOPAUSE
240
200 

STRATOPAUSE
160
120
STRATOSPHERE
40 
TROPOPAUSE
" TROPOSPHERE
3W0
340
380
430
480
500
!40
TEMPERATURE ( RI
Fig, 3.1
TM9
06
L:iI
I
ccJ
LL.
66 
0,5
70

1.0
2.0

 .. .
. .
e 0.1%
50
:40
S30
STAi",ARD
TROPICAL TE
20
PERATUR
,
1962 STANDARDTEMPERA
10N
o
60
,I
.40
,
.20 0
,I,
20 40
,\i
60 so 100 120
i.
DEW POINT (0 F)
Fig, 3.3
67
in L0ca
IS
m ccI
(0
M0 f(0 M0 4.tr
t
 in
In
al
1 0 if
~lMC M
"II
('I CV M 0 to 1 o)
4
4'4
M
toe
zt
+ +
+++
t,
C;. ic
6cl
NN L0 0)
ell coc mC 
~
In
o
4
4
4 I
W ca a.0(0 to M..t4N..i.4.~ m
.. i.. M
NI
'10 44 0
r'C4
~in
M)
tHol"0in
04 WInN
4.
.
00
u
I
fWI JOINS 4OPRINS UNIVBPSIBT
dT Td
(feet) 0.000 36089.238 65616. 750 104985. 875 154199. 375 170603. 625 200131.188 2591B6. 313 291153. 188 323002. 625 354753. 250 386406.125 480780. 813 512045. 938 543215, 188 605268. 375 728243. 750 939894. 688 1234645. 000 1520799. 000 1798726, 000 2068776. 000 (0 R) 518. 670 389,970 389. 970 411. 570 487, 170 487. 170 454. 770 325. 170 325. 170 379. 170 469. 170 649. 170 1729. 170 1999. 170 2179.170 2431. 170 2791. 170 3295. 170 38B9.170 4357. 168 4663.168 4861.168 (fR/ft)
(slug]ft3) 2, 37680982E03 7. 06119929E04 1. 7082000OE04 2. 56609928E05 2. 769 79 972 E06 1. 47349965E06 4.871 89993E07 3. 88259984E08 6. 1507989BE09 9. 65109992E10 1. 907099B)E10 4, 3, 2. 1. 8. 3. 6. 1. 3. 9. 3. 72659967E1I 56239951E12 24879924E12 55919982E12 43449DN4E13 03459948Eh13 95599719E114 26079973E14 05989996Z15 00309B37E16 04629894E16
3, 56619991E0:3 0, OOOOOOOOE+00 5. 48639B00E04 1. 536199B2E03 0. OOOOOOOOE+00 1, 09729986E03 2. 19459995E03 0. OOOOOOOOE+00 1. 69529999E03 2. 83429981 E03 5. 68669662E03 1. 14439987E02 8. 63499939E03 5. 77439740E03 4. 06099856E03 2. 92699994E03 2. 38109985E03 2. 01519998E03 1. 63550000E03 1. 10109989E03 7. 32979970E04 0. OOOOOOOOE+00
'The temperature gradient is applicable to the altitude hand which the gradient is listed and for altitude at the beginning altitude. highest next ending at the
I'
,69
4, EQUATIONS OF MOTION
Precdingpageblan 71
21
TI JOHNS HOPKINS UNIVRIrIIIITY BINVIP SPRING. MAVL6AND
The motion of a reentry body during ballistic flight is affected primarily by five factors: 1. Gravity,
2.
3. 4. 5,
Earth rotation,
Atmosphere, Body dynamics, Asymmetries,
Each factor is considered in simplified form to show the general effects on the trajectory.
',.,
73.
I I I I I
YNI J@44NM@9WN1UNIVSNUIYV
I I
WA
I
4*1
I.
t 5 L #1
75
Fr
A. ,
SYMBOLS FOR SECTION 4.1 Symbol g I K R r r t V V Definition acceleration of gravity constant defined by Eq. (4.115) constant defined by Eq. (4. 15) range distance from the center of the earth to a point on the trajectory r/r time velocity /VI r 0 path angle central angle, measured from apogee degrees radians or degrees seconds ft/see feet or nmi feet Typical Units ft/sec2
Sflight
6
Subscripts a e o refers to conditions at apogee refers to conditions on the surface of the earth refers to initial conditions
77
The effect of gravity only on the trajectory of a ballistic missile is considered. Equations that define the trajectory (velocity, time, flight path angle, altitude, nnd range) in terms of initial conditions are presented, Plots are given which show the relationship of range and flight time as a function of initial velocity and flight path angle for minimum and nonminimum energy trajectories. Also shown are plots of range sensitivity to initial velocity and flight path angles, An example of the uses of the equations is given in Example 1, Section 4.6. The major conclusions are: 1. A ballistic missile follows an elliptical flight
path.
flight path angle that results in maximum range, and the trajectory is called a minimum energy trajectory. For minimum energy trajectories, the flight path angle decreases from 450 for very short range to approximately 320 for a range of 3000 nmi; the velocity increases from 0 ft/sec at zero range to approximately 20 000 ft/sec at 3000nmi range; the total flight time increases from 0 seconds at zero range to about 1300 seconds for 3000timi range; the trajectory apogee increases from 0 nmi at 0 nmi range to about 550 nmi at 3000nmi range (a maximum apogee of about 700 nmi occurs for a range of 5500 nmi and a flight path angle of 22. 5O). 2. Trajectories for which the initial flight path angles are greater than those for the minimum energy trajectory are called lofted trajectories; trajectories for which the initial flight path angles are less than those for
weINSP4
3. The range for a minimum energy trajectory is extremely sensitive to initial velocity (about 2800 ft/ft/sec at 3000nmi range) but is very insensitive to initial flight path angle. With Increasing loft at a range of 3000 nmi, the sensitivity to initial velocity decreases but the sensitivity to Initial flight path angle increases, For delofted trajectories at 3000nmi range, the sensitivity to both initial velocity and initial flight path angle is greater than that for a minimum energy trajectory,
II
ii.
80
SI* .
VO
0V
The trajectory of a body with initial velocity Vo at ro and fired with a flight pathyo may be obtained by applyIng the laws of gravity, conservation of energy, and conservation of angular momentum as defined by Eqs. (4. 1 1), (4, 12), and (4,13), respectively. goro 2 Vo 2
2
gr 2 V2 t
(4. 11)
~
0
gr g
0 0
gr
r V cos0
00
rV cos
V2 for
0 0
81
.I
V2
2goro >
00 0 0
V2
0 1. For ballistic vehicles, 2goro, 0 is always less
0 0
than I so that the trajectory is elliptical. Using Eqs. (4. 11) through (4. 13) and the properties of an ellipse, it may be shown that the trajectory is defined by the equation:
 2 cos2
0 I 0 K cose 8,
!
(4.4)
where
r Kr i
.
0
2
K  I  VE
equation:
14
byo sti
(4.5)I
The range is the arc length OB and is given by the R 2roe (4.16)
.,
'
Thus:
!
Cos 19 =
Cos2_ o, . . .
or
"
tan
'N
(4. 17)
"82 
ii
'
It should be noted that the equations derived above are general insofar as the reference nltitude, re, Is (oncerned, That is, ro is not necessarily a surface coridition. Hlowever, if rv is not n surface condition, care must be exorcised in evnlunting the icference vLlodty, V79g 0, and It . In terms of surface conditions:
'r
(r 0 0 )
, 26 000
0 O0
(4. 18)
(4n
2r.6 R~ e
,0
From Eqs. (4, 1 6) and (4, 17), it is observed that a given range may be obtained for varioua combinations of However, there is one combination of Vo atid Vo andy , *o that resuilts in minimum Vo. This trajectory, called the minimum energy trajectory (MET), represents the minimum booster impulse forx a given range or, alternatively, represents the maximum range that may be ob
Jectory is obtained by operating on Eq. (4. 1.7) using standard procedures for obtaining maximum or minimum values, The results are the following:
or * y0
2
V0
70=45 T+:
tan
min
(4. 111)
It is noted thnt '/ decreases from 45' at very short range to 00 at maximum range (R/ro ft). The velocity gr. at maxinmum increases from 0 st zero range to Vo range. The latter velocity is the velocity for a circular
L
!+
[I
83
orbit. For ranges that will be considered in this report (R1 rr.Jqs. 1000 to 3000 nmi) the values of y and V0 computed from 42. For 41 and in Figs. (4, 16) and (4.17) are shown
all examples given in this section, ro
=
2 re
2.0 9 x 107
feet,
The time of flight may be derived in the following
manner. The rate of change in 6 expression: is given by the f'ollowing
Cos
r
(4,112)
:
u
The negative sign is required since, for convenience in deriving the equation of the ellipse, 8 is measured from apogee, Using Eqs, (4. 13) and (4, 112) the equation for time, becomes:
r
t
VCos
r d0
(4,113)
Substituting for r from Eq, (4, 1 4) and integrating gives the result:
ts
(V 0 cos '0)
(4.114)
where I
In I
Vo
2
tan' Ko
 Va
Cosn o,
 I (9/ 2 )\
(4.]1 15)
'
I K
e and So,
If one is interested in just the total trajectory time, Eqs. (4. 114) and (4. 115) may be simplified by noting  84 
'Ii
I I t !il
'
ll
l ' l......l l,
l'
that the trajectory is symmetrical about E equation for the time becomes:
0, and the
2V g
Co
...
K sLn8
+ ta
nmi in Fig. 43. For a given range, time increases with increasing initial loft angle, /,. The minimum time for a
range from 0 to B is that computed for a circular orbit trajectory from 0 to B; that is yo = 0, Vo  gCro, and t 692 seconds for R = 3000 nmi. For a minimum
tar
t 2 1J go 2
tar &
0
Cos 3
[tan a y2
"
ol
tan'
(4.117)
dr
tan " (4.118)
I6
""8...
SPINGl
MARYLANO
Solving for dr/d8 from Eq. (4. 1 4) and substituting into
Eq. (4. 11 B):
tan
= 1
(1  K cos 6)2
r .
Evaluating Eq. (4. 120) for ccnditions near impact, it may be shown that # is very small for 1000 to 3000 nmi
trajectories. Therefore, in the absence of an atmosphere, the flight path during reentry is nearly a straight line. This fact will be useful. later when we consider the effects of the atmosphere. The velocity at any point along the trajectory may
*1
be obtained by solving Eqs. (4. 11), (4. 12), and (4, 14) to obtain the following equation:
=
V
V
(4,.121)
(4.121):
At apogee 6 
 VCos 2 27
00
I K
SV V
I K cosy
"8
86
TH4JOHN$
SILVSP
HOPKINS UNIVEREITY
SP~ilNO MARYLA~N
orl~t.
o
(sin
1,
(4.122) (4.123)
" Cos
0
or
"a o
cos ^/ + siny
(47124)
For a minimum energy trajectory, the altitude at apogee is maximum (ha %; 700 nmi for ro = re) for Y. 22. 50 (R o 5500 nmi) and varies from about 250 nmi for R = 1000 nmi to about 550 nml for R = 3000 nmi. Since the sensible atmosphere extends to about 50 nmi, most
of the trajectory occurs in the exoatmosphere.
The ratio
of velocity at apogee to initial velocity decreases from 1 at zero range to 1/Vfor maximum range. For ranges of 1000 to 3000 nmi, Va/Vo varies from 0. 71 to 0.72. Another consideration of interest is the sensitivity of range to the initial values of VD and 7 0 . These sensitivity values may be obtained in the following manner:
Wvo ae 0 av0
(4.125)
aRi/6.6 0 and~60 /av may be obtained fromn Eqs. (4. 16) and (4. 1't) so that:
24)0 2 0o
(F
V sin 2oy K 2R K
(4.126) ,
87II'
V
3
(cos 2Y
 2 2 V cos K2
vY)
(4.127)
T.
Plots of Eqs. (4. 126) and (4. 127) are shown in Figs. 45 and 46, respectively. Note that the range of a minimum energy trajectory is sensitive to initial speed but not to initial flight path angle, With increasing loft angle, the range sensitivity to Vo decreases but range sensitivity to Yo increases.
88
"
3600 p
p
~I
300
450 600
3200
2800
2400
12000
1600
1200
800
400
10
I 12
,[ 14
I
16
I
18
I___ 20
__
___
_ ..
22
24
26
V0 (thousands of ft/tao)
FIg.41
89
qI
yma
uNftlvilTv JOiqNBMlOWIN@
51I.6VINpmINO. MARYLAND
70
1O0nnil
1500
2000
2500
3000
S650

40 '
MINIMUM
40
*ENERGY
10
12
14
16
le
20
22
24
26
Vo (thousondi of ft/iecl
Fig. 4.2
90
"
:1I
3000___I
2600I
1200
1800
~
40
_
so 70
FIg. 4.3
3000 NMI
91
MAIYLA.D
IS
_ wU
LJ. 9.
'Iuom
I.
A PPLIED PYISLABOCRATORY
II 1
I>
~,LL
/c
\. ,/93
I
I
'il.
/
.
22 ..
944
ww
2
1LL.
o o
i9
i __ ... L_ L
..J Ik
4.IFET FERHRTTO
I9
Symbol CA h I L R r r S t e=
Definition axial force coefficient altitude refers to the impact location refers to the launch location range from L to I re + h earth radius = 2.09 x 10x 3441 reference area total flight time it increment in reentry time
Typical Units
feet or nmL
V W
flight path angle in inertial coordinates (positive for launch angles) latitude, positive for N. latitude longitude, positive for E. longitude earth rotation angle during time t azimuth angle, positive direction is
0
A AA $
degrees
degrees radian
 9'7 
. !
Symbol Se0
Definition half the range angle for rotating earth (Eq. (4, 217)) earth rotation rate = 0. 729 x 1
10~~4rase
Subscripts e I L refers to quantities measured relative to rotating earth refers to impact condition refers to launch condition
II
I.I
...
N
.
98 
SLnVIER SPIPNO
MARVLLNI
The equations of motion given in the preceding section ire acpplic('ble to the rotating earth case provided the trajectory is dlefined in the inertial axis system. There.fore, initial conditions must be computed taking into (onsideration the velocity of the rotating earth. In addition, the amount of earth rotation during the tine interval from launch to impact must be accounted for. In this section equations are given tor evaluating the magnitude of these effects and an example is given in Section 4, 6, Example 2. The major conclusions arem I. Compared to a nonrotating earth calculation, the effect of a rotating earth on a missile fired eastward is to increase the initial inertial velocity and to decrease the initial flight path angle, Westward firings result in a decreased velocity and an increased flight path angle. The largest change in velocity (about 1500 ft/sec) occurs for launches at the equator, for eastward or westward firings, and for very low flight path angles. 2. Compared to a nonrotating earth calculation, the effect of a rotating earth on a missile fired eastward Is to increase the range and flight time. The impact latitude is the same for both calculations. For a westward firing the range and flight time are decreased.
" jj
.I A
99
fhI nS09 
?HIK JOkNBl
IO~MUI*I UNIV([RUIT'
The three basic laws used in Section 4. 1 to define the equations of motion are valid for a coordinate system fixed in space (nonrotating earth). Two major effects of the earth's rotation on the flight of ballistic vehicles will be considered. 1. The launch site has a welocity caused by the earth's rotation so that launch velocity and launch direction are modified compared to the nonrotating earth solu
tion.
2. The earth rotates while the missile is in flight, thereby affecting the location of the Impact point. Assume that a booster imparts to a vehicle a velocity Ve at an azimuth angle Oe and . flight path angle ye, where all three quantities are measured relative to a rotating earth. The earth geometry required for this study is shown in the sketch, page 102.
A
The vehicle will travel in an orbital (or trajectory) plane that contains the velocity vector in inertial space, V, and the center of the earth. The characteristics of the trajectory are defined by the equations derived in Section 4. 1, provided inertial quantities are used. The inertial quantities may be obtained from earth reference quantities as shown below. The velocity vector is divided into three components: Radial component North component East component V sin y  V cos y cos V cos y sin
i
.J
101
10
1
'U :"", "' ''
.,
1li
MERIDIAN.
AL
WEST
EAST
li
Af EQUATOR A
I/
, SOUTH
J .
In inertial coordinates the radial component and north component of velocity are the same as those measured in the earth fixed axes, but the east component is increased by we r cos 0. Therefore, the components of the inertial velocity, V, are: (4. 24) Radial component = V sin Y e e North component = Ve cos East component = Ve cos e cos Ie e sin (4.25)
102I''
THS JOHN$
HOPKINS
uNI~eftrrv
The azimuth angle, flight path angle, and velocity i~n inertial coordinates may be obtained from~ the following eqinatiuns: tn =eastward component of velocity tan orthard ompoentof velocity tan~ tan 1+ ~
w6 o
V Cosy a/in 0 e e e
(4. 27)
~.[l+2cosBY
e
sinl
~Ler~
__ e __
+ (we/ cs
a
(4. 29)
For launch velocities of interest, the velocity ratio is given by the approximation: V~ jV=+ cosY I/ si ee we 0. 729 x 10 r cos0 e Ve (4.21l) (4.212) (4.213) (4.210)
~rad/sec
From Eqs. (4. 28) and (4. 210), it is observed that 0* to 1800, V/Ve > 1 and tends for eastward launches (j to be maximum for launcties from the equator (00)for
j li
103
Tninimum energy trajectories (minimum Ve), for direct east firings (Pe = 90), and for low flight path angles (le. 0). For eastward trajectories the inertial velocity is increased and the flight path angle is decreased (compared to the values for a nonrotating earth); for westward firings, V is decreased and  increased (compared to the values for a nonrotating earth). The maximum variation of V/Ve from unity is +er/V . For V z 10 000 ft/sec the maximum variation in VfVe is about 15%1, Given values of Ve and YeI the inertial values of and Y may be obtained from Eqs. (4. 28) and (4. 210).
in Section 4. 1. However, the range computed by Eq. (4. 1 9) corresponds to the range measured along a sphere fixed in space. The range measured on the rotating earth (in the absence of an atmosphere) may be computed as shown below, The orbital plane defined by V and the center of the earth remains fixed in space. During the flight time the earth rotates through an angle:
AA = 57.3 w e t
(4. 214)
For given Inertial launch conditions, the impact latitu1e is unaffected by we and, from spherical trigonometry,
sin 0,
where is the value computed from Eq. (4. 17) for Yo0 and V. = V. The impact longitude is given by:, AL  57.
3
0"et + sin"
o
(4.216)
where t is the flight time computed from Eqs. (4. 17) and (4.116) for Vo Vando Y,
104
t
I.
T"7 J
l
01, and A, the range may be comGiven 0 L, A puted from the equaton:,
R = 2re9 e e (4, 217)
From the spherical geometry shown in the sketch, the range angle 2 6 e may be obtained from, o 2 cos 01 cos 0 cos A+ sin 0 sin
0
L '
(4.218)
*1 .1
MERIDIAN EQUATOR A
AA
The bearing of the impact from the launch Atte may be obtained from cosA co Cos cos 2e cos
e
COs SLI sin
26
e sifL
(4.219)
S
105
105i
Ti
For
sin 0I
Cos
_ _
(4. 220)
COS
CBAA Cos
Id
Sometimes V, /, and lare given and it is desirable to obtain the corresponding quantities relative to a rotating earth. The required equations are:
v 2 Ve =[Ii _2er
Ye
cosyV sil V
(~ + .WO
Cos 2 0
11/2
(4.221) (4. 222)
sin Y
V
WeV cos 0
e
tan er
tan
V cos Y sin
(4. 223)
16
'
ij J
i
I I I
~~~.
I I
13
I  107
'
Definition
axial force speed of sound axial force coefficient
Typical Units
pounds ft/sec
Cmq CN CN D d
g gX
pitch damping coefficient normal force coefficient dC /d& damping factor (Eq. (4. 520)) reference length
acceleration of gravity longitudinal acceleration
feet
ft/sec2 g
H
h h max
feet
feet feet
" I (K) In
1 2 (k) I
(K/2)n n.l
(2K)n O 1.3.5...(2n+l)2 n,,0 slugft

nol
e K 0
pitch (or yaw) moment of inertia parameter defined by Eq. (4. 36) constant defined by Eq. (4. 326) reference length Mach number  V/a vehicle mass atmospheric pressure
K k1 L M m P
feet slugs
lb/ft 2
109
UINIVKRIITV
,9ymbol
q qs
Definition
dynamic pressure :
rypicai Units
Z PM 2
lb/ft2
stagnation heating rate per unit area total heat per unit area absorbed at the stagnation point gas constant s 1716 Reynolds number, VL/v vehicle nose radius reference area atmospheric temperature time intervals of time defined by Eqs. (4. 320) and (4. 321) vehicle velocity vehicle weight distance from vehicle nose to the c. p. distance from vehicle nose to the cg. X
Q
R 1k RN S T t At1 , At 2 V W X C. p. X c.g.
feet ft 2 OR seconds
2I
4X
x
[3ballistic
y
V
radians lb/ft
reentry flight path angle (always negative) kinematic viscosity atmospheric density
WA
110
I
T4I9 JMNNS HOPKINS UNIVrADITY
Sl~~it.vS ~ii
Subscpts. ave E 0 refers to average over the time interval refers to reentry conditions (K=O)
refers to initial conditions
max
A dot over a symbol means the first derivative with respect to time.
.1
'I
I.
II
*1
~
111
.....................................
*"...,n
.i.
SpPtNQ M~mA~N~h
SUMMARY
vehicle wovw s throlug(h nn rintm( . phiree, aerodynajmic for'ces nt. genorflte( thAt ;1il'ced. Velicilue notien, The magnitude ul thius Iu U'LUB is L11lUCnlCUd strongly by the missile velocity and atmospheric density. In this section approxinate equations nre derived fo;: missile velocity, ambient density (or pressure and temperature), and related parameters.
When i
Assuming that the axial force is the only force acting on the missile, and making other appropriate simplifications, equations are derived tor velority, time, dynamic pressure, longitudinal deculeration, Reynolds number, stagnation point heating rate, and the total heat absorbed at the stagnation point as functions of a parnmeter (K) which, for a given missile, atmosphere, and trajectory, is defined by altitude only. Most of these quantities increase from zero at reentry, reach a maximum value,
A*1
and then decrease prior to impact. Equations areoccur. given which they for the maxima and the value of K at An example illustrating the use of these equations is given in Section 4. 6, lExample 4.
13
4
I l4~k
rii
JOINM,
MOffINB UNI~t~gtY
The boot;t phase of the trajectory is guided to such a high altitude that atmospheric effects during the exit phase are usually negligible, Therefore, this discussion is limited to the reentry phase. Custo.narily, reentry into the sensiNl ntmosphere is assumed to begin at an altitude of 400 000 feet. The velocity increases slightly under the effect of gravity, reaches a maximum value, and then decreases under the influence of aerodynamic drag, The maximum velocity occurs when the axial force is just equal to the component of gravity along the flight path, or:
C q S W sin y
AI
h)
VE
max If the reentry velocity, VE, is given for some altitude above that corresponding to Vmax, the maximum velocity may be estimated using the equation: i"
Vmax VE + k
where k is a constant having a magnitude between 0 (for ) A precise value of hE f hmax) and I (for hE ",> h k is not important since the Jflrence between VE and Vrnax is small and, In fact, is usually neglected. However, if a little more accuracy is desired, the equations for p and Vmax may be solved by successive approximation. The remainder of this section considers the motion oi the vehicle during the deceleration phase. During this
LJNIVEMSI
period the general behavior of the body may be obtained by simpli[Ld equations of motion based on the following assumptions: 1. 2. The earth is flat and nonrotating. The only force acting on the body is the axial orcu, and the axial force coefficient is constnnt. The reentry flight path angle is constant.
:3.
The reentry phase is of relatively short duration and short range so that the assumption of a flat nonrotating earth is justified for this study. The assumpti.on that the axial force is the only force acting on the body implies that the angle of attack is always zero and that the effect of gravity is negligible compared to the effect of axial force. More precise methods of computing trajectories may be used to show that the effect of y Is usually small, Except for the veryhighaltitude and, sometimes, the verylowaltitude portions of the reentry trajectory, the assumption that
W sin V
A
mate solutions to the equations of motion, Precise methods of computing trajectories may be used to show that assumption 3 is valid for a large portion of most trajectories and that the validity tends to be improved for high weight and low axial force bodies and for steep reentry angle trajectories. For many reentry bodies and trajectories in current use, the assumption of constant V' is valid all the way to impact. The geometry of the trajectory in which A is the only force acting is shown in the sketch.
A)
S~~~
1161
1
I
IN& JOHNS HOPKINS U"Iywsl3m y
A HORIZONTAL
7
V
IS NEGATIVE
DURING REENTRY
hI
R.NGE
The equation of motion along the flight path is: A 'dv (4. 31)
CA (g 2)pV S
WdV
T 
(4. 32)
/1 V sin ^
Since the flight path is a straight line, dh/dt and Eq. (4. 32) may be written in the form:
dV
V
CagPS dh
2W sin Y
(4. 33)
The term W/CAS occurs frequently in reentry equations and is called the ballistic coefficient, A. For an atmosphere in static equilibrium (see Eq. (31)), dP = pgdh. When these substitutions are made, Eq. (4. 33) becomes:
IN
117
.I
APPLIED PY6ICS
mILVIN 59n140
LABORATORY
MAiPTLNNV
dV _ dP 203 sin Y V
Using initial confitions V Viat P
(4. :34)
Pi, Eq. (4. :34) Is
(4. 35)
P
KV
2
(4. 36)
eE
 << K
This con
listed above. dition is a restatement of assumption 2 invalid at low rebecome to tends Therefore the equation high values of K. entry velocities and at very low and very is not serious and The limitation at very low values of K section. On the was discussed at the beginning of this of K may have other hand, the limitation at high values 2 for example). 1. significant consequences (see Section 5. 0 and the equations When the initial altitude Is high, K.i of these equamay be simplified, For some applications altitude condition. tions the initial condition is not a high is retained in Therefore for completeness the K1 term Eq. (4. 35) and subsequent equations. may Other trajectory characteristics of interest Eqs. in given are be derived (Ref. 41) and the results from altitude hi (de(4, 37)through (4. 329). The time by P) is given fined by Pi) to any other altitude (defined by: ave
I
t t.
_ In +
I
I (K)
K(4.
37)
S118
I
?iii
where the I1 function is plotted in Figs. 47 and 48. dynamic pressure is given by the equation:
The
q:2 C AS11
2 C
si
eK  Ki
(4. 38)
KV
sin Y
c KKL
gx
(4. 39)
The Reynolds number based on body length is given by: ViL ve (K/2Ki/1 (4,310)
The heating rate at the body stagnation point is indicative of the severity of the heating environment. This rate is given by:
412O)j
e 3 (Ki
K)0
(4,311)
sn ~*
12 (K
12 (K)
(4
2
where I 2 (K) is plotted in Fig. 49. The dynamic pressure, longitudinal deceleration, Reynolds number, and stagnation heating rate all have a
119
JOHNS NOIKINII uilVEIlrtY MTI APPLIED PHYSICS LABORATORY *1L, VlN SPRING MA NI1.1&nI
trend which increases from zero at reentry to a maximum value and then (depending upon the body and trajectory characteristics) decreases. The approximate pressure at which the maximum values of these functions are reached mny be obtained by assuming T is n constant, KL = 0, seLting the derivative of the f'unction with respect to K to zero, and solving foe' the maximum, The error introduced by assuming T is 0, VI = VE and the equations a constant is small. At K1 for the maxima and their X, P, and V/VE values are shown below, Equation
K
1
P
0l sinY
F.
0.80Oi (4,.313)
S"
q
2 V i
s~an Y
0,606
(4,3143)
gX"9330T
!R
I,.
0', 8
2 nn
0.8
435
/ 8 o20(0/3)
T16) 1/0
sin'
0,646
(4,
The longitudinal acceleration is a characteristic of considerable importance since it is a quantity that can readily be measured on a flight test vehicle. It is interm esting to note that for an isothermal atmosphere is highly dependent upon VE, moderately dependent upon y and T, and independent of the vehicle weight and axial force (p). However the altitude at which The greater the 0 or the. occurs is dependent upon , lower the altitude for the steeper the reentry angle, gXmax'
'1
Ii,
120
'C
"'a65 JOHNI$
"IOPkINI UNIVII1if1?
(4. 317)
.
(4.318)
!ind

E ..
2g(H)2
K(s
K)e 3K/2
(4, 319)
Sg9x
or minimum. and K  2,
respectively, or P = (1/3 )g sin / and 2g sin Y, respectively. Using Eq. (4. 37), the time between the high altitude inflection point and gXmax is given by: 80T
At 1
ave sin Y
(4.320)
and the time from gXmax to the low altitude inflection point is given by: 77T ave V sinY
t2
'
(4.321)
121
_
_I 
.
TM9 JOHN$
Sli..* Vi
NOPRING UW IrMITU,
*, lN6 M&AALAND
14)(
2 (4. 322)
g 
gx
' 2. 72
Therefore, using just the telemetry gx trace, the ratio of gX/gXrnax is a good indication of the K history. The K history can be converted to air pressure history from known p and 7 (Eq, (4, 36)), and the pressure history can be converted to altitude history from known atmnospheric characteristics. A plot of gX/gXMax versus K is given in Fig. 410. Other quantities required for succeeding sections are listed below for convenient reference,
2
WA
K
eK (4,323)
1/2
kA 2
tir
A E
D
CAIYI a H
]
]k1/T2
(1
(4.324)
(4.325)
Ix
ly (4.326) :
=C
 122
__
q
2 V 3m sin7
(1 K)
e
'K T2/
(4.327)
2CAS H
V2
K
E 2 2CABH rn (,y)
(0 eK  K)
(4. (, 32B) B Sd
A
V
2 WVsinn y
CASH
A
77143
e K(4.329) E
The 1962 Standard Atmosphere characteristics required for evaluating these equations are listed in Table 31.
 123
I.
f0.J
.;0
0.7
11(K
IK 2
0.6
nI 0.2*1
nni
'
0.1
0.2
o.
04
0.6
0.8
1.0
1.2
.L
1.4
Fig 47
125
MARY6ANG
12
IN
1261
5A
lH4
JOHNI HOPKINS
.JNIVIONIIY
1,0
...
0.9
0.7 0,B
0.4
0.3
FIg.4.9
127
___
*
*
~APPLIED
PH'YSICS I.ABORATORY
MAIMANO
*~tV3~SPOI140
I...
LI.
m*
C1
0O
0e
128
I. I.
Is.
:}
12
I~
129 
I
?HS JOHN
SILViE
Symbol a C1,C2 C 3, C4 , C CA
Definition constant defined by Eq. (4. 452) constants in Eq. (4, 433) onstants defined by Eqs, (4.445), (4. 458), and (4, 461) axial force coefficient normal force coefficient
Typical Un its
CA
CN
C
N
ddCN/dF
N
C1
p Cm d FXi Fy, F F g gL . q
aerodynamitc forces along body axes X, Y, Z CN qS acceleration of gravity load factor = N/W scale height v RT/g or angular momentum
I0
PIx YIy J K k1 k2
axes X, Y, Z
zero order Bessel function P/(A sin>') Y); constant defined by Eq. (4.430) constant defined by Eq. (4. 431)
slugft
rad/sec: rad/sec
1131
o'w
Symbol 1, ni,1n
Definlition
Typical Units
aerodynrinniic rolling moment, pitching moment, nid yawing moment, respectivelv mnss  W/g roll damping factor, C a
m M
'1
'4
I2
V 2
qSd2 V
lbft se
2
lvi Mr yaw damping factor, C Vnr 2 + F lbftsec
N P p, qj r q R S T t u, v, w
roll rato, pitch rate, yaw rate, respectively dynamic pressure gas constant = 1716 reference area temperature tirn.e missile velocity along X, Y, Z axes, respectively
2/sec
ft 2 GR seconds ft/ see
V
W X AX Xg.
missile velocity
missile weight constant defined in Eq. (4. 440) X c. p.  X e.g. distance from missile nose to the c. g.
ft/ see
pounds
feet feet
F'A2 C. .
Ilk.r
Ii
tZe JOMN 4OPIMW UNIVINGIIlv APPLIED PHYSICS LABORATORY
Symbol X ry c.p.
Definition distance from missile nose to the c, p, angle of ;tt, ck in the XZ plane
_ne2 +
radians
radians lb/ft 2 degrees degrees degrees degrees rad/sec rad/sec rad/sec rad/sec rad/sec seconds seconds degrees
angle of attack in the XY plane or ballistic coefficient, W/C AS flight path angle, negative for reentry trajectories ( angle defined by Eq. (4. 422) defined by Eq. (4. 423) angle between the missile Xaxis and the local horizontal g WT W A frequency defined by Eq. (4. 421) transverse rate =Vq + r
0angle
2 +qr + total spin vector = aerodynamic frequency defined by Eq. (4.436) defined by Eq. (4.449) period defined by Eq. (4. 424) period defined by Eq. (4. 425) angle
Sfrequercy
"B T
Sphase
Subscripts E o
A dot over a symbol means the first derivative with respect to time. Two dots over a symbol mean the second derivative with respect to time.
133
In the preceding suction, the effects of the atmosphere on vehicle velocity and related factors were cons idered. In the present section, ittention is given to the effects of the atmosphere on the dynamic behavior of the vehicle, We are concerned here with the rotational motion of the vehicle about its center of gravity over relatively short periods of time. Sixdegreeoffreedom equations of motion are presented and interpreted in terms of body motion characteristics in the absence of aerodynamic forces (for example, motion in the exoatmosphere, where gyroscopic motion is importani) at low altitudes where aerodynamic forces dictate the motion and at high altitudes where aerodynamic and gyroscopic forces are both important. Dynamic motion is important in the analysis of body heating and lateral loads. Equations are given for body rotational frequencies that are observed on telemetry traces from onboard equipment or from tracking radar signal strength traces. Numerical illustrations are given in Examples 3 and 5, Section 4. fJ. The major conclusions are: 1. In the absence of aerodynamic forces, a reentrytype body that is spun about its longitudinal axis will follow a motion pattern determined by conditions at the initiation of the spin. If spin is applied so that neither pitch nor yaw rate is introduced, the vehicle will maintain a fixed attitude in space. If a pitch and/or a yaw rate (sometimes the vector sum is called the transverse rate) is experienced when spin is introduced, tUe missile's longitudinal axis traces a cone whose apex is at the missile center, of gravity. The magnitude of the corne angle and the period of motion are functions of the transverse rate, missile roll rate, and moments of iertia about the longitudinal and transverse axes.
 1 135 
THE JOHN*
iOPtOINI UNIVaIFITY
2.
3.
phase, aerodynamic forces dictate the motion of the vehicle. Generally, a missile reenters thr, atmosphere with some initial angle of attack. To prevent excessive lateral loads and heating problems the missile is designed so that the aerodynamic forces cause the angle of attack to approach zero rapidly. It is shown that for a symmetrical missile the angle of attack follows a damped oscillatory motion provided certain static and dynamic stability criteria are met. The frequency of the oscillation is called the aerodynamic frequency. This frequency increases from zero at reentry, reaches a maximum, and then (depending upon the trajectory and reentry conditions) decreases until impact. 4. At high altitudes, both aerodynamic and gyroscopic moments are important. The gyroscopic moments tend to maintain the body at a fixed altitude; the aerodynamic moments tend to rotate the body so that zero angle of attack is maintained. Typically, gyroscopic motion predominates at altitudes above about 300 000 feet; aerodynamic moments predominate at altitudes below about 100 000 feet; both effects are important at the intermediate altitudes,
136
?Ir
JOlNP4
MOPKINS UN.V3Nt9 ?Y
The primary interest in this section is to consider how the body behaves over a relatively short time span. For this purpose the effect of gravity may be shown to be very small. By assuming the body to be symmetric about the longitudinal axis, the principal axes coincide with the body axes, and the equations of motion are simplified since the productofinertia terms are zero, The equations of motion are presented in a body axis system (Fig, 411) which remains fixed with respect to the body. The body rate terms In the equations of motion are the rates sensed by Iate gyros mounted in a vehicle; the body translational acceleration terms are those that would be measured by accelerometers mounted at the vehicle center of gravity, The six equations defining the six degrees of freedom are: NX = I  qr(Iz  I 4I = m pr(Ix
 I)
1X
'IZ = n pq(Iy
7,
F S 9
y
+ vrwq
X
(4.44)
137
I,
= Fr g + +wp
ur
(4.45)
Fzg
+uq  vp
,
(4.46)
(4. 47)
(4.48)
n a FyAX + M r y
u 2 2
(4.49)
V:U
+V
+w
(4, 410)
_ Vv2 +w2
tan; FX = CA'S Fy  F 2"v F,( (4.411) (4.412) (4.413)
IV2 + w2
F7
F
(4.414)
%V +w
4CLqS
9s
(4.415)
138 
Zall
JOHNSMHOPKINS
UNIV90111"
4.4.1
Consider first the case of flight in the exontniosphere where nil nerodynamin terms are zero. In this region only gyroscopic effects exist. Since the body is assumed to be symmetrical, Iz  1y and Eq. (4. 41) reduces to the condition where p equals a constant, Differentiation of Eq. (4. 42) gives:
,4
+ pP
Iy
(4,416)
q = 0
(4.417)
COS CoT
T II
p(l  Iy) t +t
qmax
then
)
(4.418)
Y
SW OP(I r)
sin
p(1
IX

139
I
S'
" ' "I  ',,. , . .
'  1 . .
t 2
=W
Therefore, the pitch or yaw rate measured by a rate gyro mounted on a vehicle is a sLnumoidal oscillation of halfamplitude WT having a frequency:
IX p(l  Xg ,Y)
(4.421)
This equation is convenient for checking the calibration constant for the roll rate gyro in flight test work. The term WT is sometimes called the body transverse rate. The solution to the equations of motion in body axes provides information regarding the time history of quantities that may be sensed by "onboard' instrumentation, but no information is provided on the motion of the vehicle in space. The motion in space is obtained by solving the equations of motion written In the inertial axis system. It is shown (Ref. 42) that, in the absence of aerodynamic forces, the motion of the body may be described physically as a "body cone" rolling along a "space cone" (or "cone of precession"), as shown in the sketch, page 141.
140
All.
__"_____ __,___ ___"____
II
?T"e J)ING MOORPING iJNIVVINNITY
APPLI[
PHYSICS LABORATORY
SlLVIR SPRING MANVL.Ako
BODY CENTERLINE
PP
1 Y of the space cone is the angular The centerline W~ momentum vector which remains fixed in space. The centerline of the body cone is the vehiclets longitudinal axis of symmetry (roll rate axis). The semicone angle
tan =
(4. 422)
S~
141 
'I
I
TVII 10MfphlMPIgM I I UIM L~i vttIq vlllT
is constant and lies along the space cone surface, Assuming a positive roll rate and IX < I , the body centerline rotates about 11in a clockwise dyrection (as shown above) when viewed by an observer looking toward +H. The transverse rate vector rotates about the body centerltne so that the period is given by: T (4.424)
and corresponds to the frequency given previously in tEq, (4.421), The body centerline processes about the space
(4,425)
Ty p V I + tan2
For reentry bodies of current design, rs is an order of magnitude greater than rB. The utility of the physical picture described above may be illustrutted by the following example, Suppose we wish to determine the coning motion for n specified set of conditions at reentry, In particular, select a condition at t . 0 where 9 = ; and lies in the body XZ plane, which is the plane of the trajectory. Consider first the case of qo 0 and r ro, where r is positive and may be represented by a vector along the +Z body axis. Since W lies on the surface on the space cone and the body cone is external to the space cone, the angular momentum vector lies in the trajectory plane and E varies from W. (a maximum value) to 20 (a minimum value). Similarly, if ro is negetive, b varies from o (a minimum value) to a maximum value of o + 20, The bodyfixed Yaxis at t  0 is directed toward the right of the trajectory as viewed by an observer looking toward the impact point. Therefore for an initial condition of pitch rate only, if q is positiv,.
o
142
I}
p X BODY CENTERLINE
HORIZONTAL
H
JV
I
the angular momentum vector is directed to the right of the trajectory so that the nose of the vehicle remains in or to the right of the trajectory plane, Similarly, if qo is negative, the vehicle nose is always in or to the left of the trajectory plane. 4. 4. 2 Zero Gyroscopic Forces Consider next the case where the aerodynami.c terms are much greater than the gyroscopic terms (body rate terms). In addition we will make the following assumptions: 1 The trajectory region of interest is over a very short time period so that the dynamic pressure is constant. 2. 3. 4. The roll rate is zero. The pitching motion is planar (I. e. , only pitch or only yaw motion),
ey is small so that the use of linearized derivatives is valid, and cos 0 u 1; sin .,
Suppose that at some point along the trajectory the vehicle i perturbed to some a * So We wish to study the
143
'.1
resulting motion of the vehicle under the influence of just aerodynamic forces. The geometry of the trajectory Is shown in the diagram,
HORIZONTAL
0'
S"Yo
BODY CENTERLINE
, .
(4. 426) 1,
From Eqs. (4. 46) and (4, 426), using a small a approximation, and noting that u may be shown that:
.
V cos
and w =V sin;,
it
.'
'WV
(4.427)
144
1i
1T4K JOHONP MOPKINI0 UNIVtSIQI ,T
l Pq mP tIIPv MARV6IA NO
PI
Z
CN qS
qS
(4.428)
FZ  N
Using Eqs. (4. 425) through (4, 428), it may be shown that the body motion is defined by the differential equation: o'+ k 1o + k 2 ev where 0 (4, 429)
CN
Cs
wv g
kl "*W 1i
C "* 
C sd CM
q~d
(4, 430),.: I V(440
CN 6XS a 
C q
CN
(cSd)2 (4.431)
YW g
V2 Y
term of Eq. (4. 431) Is mq very small compared to the CN term, and the equation for k2 becomes: k2 CN XqS
42 (4,432)
ly
145
!i
i*
For statically stable vehicles, k 2 is positive, and the solution to Eq, (4, 429) is:
1C, sint
+ C2 cos t
2 "4
(4. 433)
where C 1 and C 2 ate arbitrary constants deiined by the initial conditions. Let No and F  0 at t = 0 so that C2 = ?o and C1 = 0 and Eq. (4. 433) becomes:
k2
cost
(4.4:34)
The cefore, for a statically stable vehicle, the n history is oscillatory and is damped (converges to ,oro) provided k1 > 0. if k1 < 0, ; diverges even though the vehicle is statically stable. Some flight tests of reentry vehicles have been observed to exhibit this type of ; divergence. To assure W convergence, then: k > 0 or AX 0 static stability2
(4. 435)
> 0 dynamic stability
k > 0 or C 1 N
Md

m qI
equation:
WA
k2
"W
k
"146
o'lt
s'RII'I
A X (IS
(
4.. 4.Ar,)
A
This frequency, called the aerodynamic pitch f'equency, is of gront importance in the motion of reentry bodies is will b. shown later. It should be nAted that is Zero :d reentr (C, = 0)and, for constant CN and W AX, W A is proportionrl to lq.
A solution to the equations of motion (Ref. 43),ln which the d.)namic pressure was allowed to vory, shows only small d[ffermenes frorm the results given above cxcept for very blunt bodies and nearly zero values of X. The aerodynamic frequency may be evaluated as a function of altitude (T, K) using equations derived in Section 4. 3. From Eqs, (4.3.8) and (4.4. 36):
2 A
4.4. 3 Nonzero Gyroscopic and Aerodynamic Forces At very high altitudes the dynamic motion of the body is governed by the gyroscopic terms in the equations of motion; at low altitudes where the dynamic pres sure is high, the aerodynamic forces predominate. We look next at the transition from one region to the other. Using the equations of Ref. 43 and the reentry characteristics given in Section 4.3, it may he shown that r 0 at reentry the ty history is defined by the for p = q equation: 147
TM& JOHN*
HOPiKINS UNIVErRII.
ry ry L
ll
(4. 4.3 B)
2C
CXPI(C
J4( A
C
CI
q Y
where Jo(X) is the zero order Bessel function of thi first kind plotted in Fig. 43 2. Writing WA in terms of K using Eq. (4.437):
2Hm

N, CA1 Y " Cm C
1 , (4. 4:39)
Jo E C c xp
(V
sin,
K [Tr
(C2 N
Iy
A
2C
qY n
Therefore, for a given vehicle and trajectory, the C! history Is a functLon of P and T only and P is the predominant factor. Even without the exponential damping function in the denominator ofEq. (4, 439), the ey history is a damped oscillation about y = 0, and the frequency is the aerodynamic frequency, WA. For conditions near entry (4 ; 1), K  0 so that the exponential term in Eq. FeE (4. 4390) 1, and writing K in terms of P:
I.
I
S2 01Eg
_
~2SF!
sin y Y /
Jo(X) .
(4, 440)
Equation (4. 440) may be used to determine the altitude (P) at which the atmosphere begins to affect the body motion. Arbitrsrily selecting vy/e.E = 0. 99 as the eondition when the atmosphere begins to affect the body motion, from the Bessel function tablesthe argument, X, corre0.99 is 0.2. sponding to F'l E 1448
'' P
4.441) (X
The altitude for the first peak in a corresponding to X = 3. 83 shown in Fig. 412 may be obtained from:
sin 2 ' 7. 34 1 g 2
Iy
SHCE AX N
(4.442)
The ; envelope could be obtained directly from tables; however, for ;/WE less than about 0. 8, Jo(X) Jo(X) is approximated quite closely by the equation:
J (X)
(X 0o
x)
(4.443)
V
sinv
e3
where
c3= 4\C A
Iy2 c A md
AA
(4.445)
At high
149
Tt41 JOHN$
"OPKIINI
UNIVIlSyT
2I
s In
~(4. 446)
The ry convergence at high altitudes is a function of body ), aerodynamic charmass ehnracteristics (I, S, X
acteristics (CNd Xc, p,,, type ofki'ajectory (YC) and ntinospheric properties (T, P). For a given trajectory, mass, and aerodynamic characteristics, a Is inversely propor
tional to
1'
or approximately toA P,
V or a given
mass, aerodynamic charactcristics, and altitude, the That is, at a given convergence is proportional to V altitude, the o for a shallow trajectory will have converged more than that for a steep trajectory. Increasing CNI and AX tends to incvease the convergence at a given altitude' increasing Iy/S tends to decrease the convergence. If p A 0 at reentry, the early ; history is changed, The gyroscopic terms tend to maintain the body motion about the angular momentum vector, which is fixed in space and usually is not directed along the velocity vector. The aerodynamic terms tend to cause oscillation about the velocity vector. As a result, in the Initial phase of reentry, i oscillates about an angle of attack which, itself, is gradually reduced as the aerodynamic forces increase. The equations defining and / for the case of WT = 0 are (Ref. 44):
S
A cos

4t (4.447)
150
I
UNIV911181 JOHNS HOPKINS APPLIED PHYSICS LABORATORY
SIl*1R
SPSNG
TiH
M&ftI.IMLA
X*~~'
2
y
Lx
y
(4.449)
p, the
OwlI
"A xP
II
(4.450)
iy
Where WA shown (Ref. 45) that the ratio a roll is given by the equation: is not small compared to
1
IX p, it may be
I1
21
itmy
without
with roll to
a tanh a &Poo
HpIx
(4.451)
"a
Eysin
(4.452)
1, and the 1
For values of a greater than about 3, tanh a a history corresponding to Eq. (4. 444) is:
151
.IL
;,, : . :' .'l..,,.. /... i. .
Y A
(4, 453)
e 3
C 3K 1, the o. enve
A A7w
J,
4
For a nonrolling vehicle at reentry, a begins to decrease when the aerodynamic moment Is sufficiently using Eq. (4.441). hlgh,and this altitude can be estimated For a rolling vehicle, re begins to decrease when the aerodynamic moment is significant relative to the gyroscopic moment, This altitude may be estimated using Eq. (4. 450) and solving for
aA AL
0. 99, so that:
WA
0.1 I Iy
(4.454)
=
RT
2
 p2
2
0.
2 AXSVE
(4. 455)
1
T
I C
Y N
A rapid convergence in a is desirable in order to prevent excessive lateral loads resulting from the 3E. The lateral load factor is given by: S gL " N =CN E Wqo(E ( (4.456)
1
,"
152
/Efrom
and i from Eq. (4. 38), it may be sgown that the maximum
23
9L E 12_'
1/4 K1 Kl elI Ri CA rd
gH
c+
A
0.75
A
(4. 458)
IV
the maxi
i4
K 3 :.
(4.459)
74
Using Eq. (4. 459) and substituting for the constants in Eq, (4, 457), the magnitude of (gL is given by:
maxI
2 0O,1 2 8"s. max
n Y)
C C
3/4
where C 52
CA + LC
A
N a4
2 m~
a:
(A
C4
(4.461)
I
?HII JOHNE HOPPINSl UNIVURI,,IT
S#~n*I
MIPIYI~k b
I
i
In evaluating Eq. (4.460), a must be in radians. E The corresponding equations for a rolling vohicle at reentry are,
2 2 31/ 4
I
462)
IL
0 42
K(4,
and
3E) 114
3 /4
N(
3/4
1/2
w0.16
(4,463)
LTI;I
The maximum vnlue of gLfor a rolling vehicle also occurs 3 1 at Ku . 4 The ratio of maximum gL for a rolling vehicle to maximum gL for a nonrolling vehicle, obtainied from Eqs.
I
g Lp= 0
at' K
/ V[il l  154
L, 1. "2
H" X 1"
'
U, V.
w VELOCITIES
mi41
BONOXS YTEM:
I OI EVLE HW
2.__tona_
You
Ul
IsI
I
~
15
?weU
~4044
UNIVISOITY wmoe~~
It
'157
I
rmi JOHN* PIOPKIND UNIVeypITY 61'VKW $04,000 MAftVLAND
Symbol A
Typical Units
a, a1l a2
CA C C C C
0
Cm CN Cn
i0
(45100a/e
bCm /(qd/V)
pitchdamping coefficient,
w0
N
C C 1 ,C c. g. c. p. D d F F
N
2
dC
/dy rad/sec
constants in Eq. (4.5110) center of gravity center' of pressure damping function defined by Eq. (4.520) reference diameter 4F2 2
feet pud
pounds A
asymmetry factor defined by
Eq. (4.572)
159
. ....
...
.. ....
. ...
I.NIVURPIYI
S ymbol FXPF F
&
Definition forces along the X, Y, Z body axes C qS N function defined by Eq. (4, 547)and (4, 573) factors defined by Eq. (4. 5125)roll, amplification factor defined by Eq. (4, 537) G required for roll "lockin"
F(X. D)
VV V2
F VF G GR GI
roll amplification factor defined by Eq. (4. 580) of gravity gacceleration gXlongitudinal G1 G 2 11 11altiLude IA 4 A Iz moments of inertia about X, Y, Z axes acceleration
ft/ sec 2 g
roll amplification factors def'ined by Elqs. (4. 5 81) and (4. 5 84) scale height, lRT/g f eet feet slugft2
1'
K
Y, Z2 1 products of inertia about 1 5101)) (Eqs. (4. 599) to (4, X, YZaxes xyX~z
P/($ sin 7)
slugft2
k, k 1, in, n
0
constants defined by Eqs. (4, 559) and (4. 560) aerodynamic moments about X, Y, Z asymmetric moments abouit X, Y, Z Mach number roll damping moment/p
I M In
0
M M p
160
AP,'L.IC
P0ipYSICS LASCRATORY
Symbol
M M
efinition
pitch (ha iutnF monent/q yiiw (1ch1pinp, vehicle Ln:SS mrioment/r
Typical
lbftsec lbftsec
UnatN
lb/ft2
S~Probtibil~ty
n, q,
nj
rdAlsec / SCc
P'rad
term defined by
SVe",
Ap
gr, js constaint r
S T t V W
1716
f' I2 t 2 'i Pcow s c pout...] j feet reet fe t feet rad ,ns radians
x
AX Yc. g, co, g.
16,..
*1O ,AA
S.mbol a
_ p
Typical Units
radians
radians dnn lb/ft2
~~anglu,
or
angle of attack resulting from 8 XY plane at in the o u co radians radians radians radians degrees degrees
ry
6inclinatinn
XY plane 6 62 2
632
ang e defined by Eq. (4. 583) y flight path angle, always reentry negativedees or ratio of specific heat, 1. 4 space cone halfangle or angle defined by Eq. (4. 531)
sin 1
degrees degrees
0 C. g.
degrees
oM
Pr/WA
A cos A.Cos ] r/d  r/d)min
degrees
OI
162
(.<
MAU4"..APSO]
Symbol
Definition
angle defined in Section 4. 5. 1 angle defined in Section 4, 5. 1
TypLca! Units
degrees degrees seconds rad/sec rad/sec rad/se:
WA
Sp(1  I /
*T Subscripts c. g. E e i M max amin o R S. L.
refers to c. g. or is caused by c. g. offset refers to reentry condition refers to equilibrium condition refers to conditions at t = 0 or to an elementary quantity caused by asymmetric moments refers to maximum value refers to minimum value refers to conditions at p = 0 refers to resonance conditions refers to sea level conditions
A dot over a symbol means the first derivative with respect to time. Two dots over a symbol means the second derivative with respect to time. An arrow over a symbol means a vector quantity.
tN
163
"41k
I
T"94 JOHNS MHOPKIN9UNIVS0ITV
I
SUMMARY AND CONCLUSIONS A reentry body is normally spun about Its longitudinal axis to provide stability during the exoatmospheric phase of flight and to reducc dispersion resulting from phenomena that occur during reentry. Ideally, the roll rate remains nearly constant during reentry. However, under the influence of mass and aerodynamic asymmetries, the roll rate may change radically. Under certain conditions the roll rate may change so that the roll rate and the aerodynamic pitch frequency are nearly the same for long intervals of time. When this occurs the roll rate is said to "lockin" with the aerodynamic frequency. This situation is sometimes disastrous since the angles of attack and thus the lateral loads become very large, The loading situation may be aggravated further by aerothermoelastic interaction. On the other hand, the asymmetries may cause the magnitude of the roll rate to decrease and change sign. In this case the missile will be driven off course in proportion to and in the direction of the lateral loads existing at the timne the roll rate is nearly zero. Even if the roll rate passes rapidly through zero roll rate, the impact dispersion may be excessive for high performance vehicles. The sixdegreeoffreedom equations are modified to include various types of asymmetry, and the conditions under which excessive loads and/or dispersion may exist are discussed. The major conclusions are: 1. The effects on body motion of aerodynamic asymmetries (abnormal pitching, yawing, and rolling moments, and center of pressure offset from the longitudinal axis of the missile) and mass asymmetries (center of gravity offset from the longitudinal axis of the missile, products of inertia, and unequal pitch and yaw moments of inertia) are considered, All asymmetries considered
 165
,TH
'IS lT
have some effect on body motion. However, for reasonable magnitudes of asymmetry, the effects of unequnl pitch and yaw moments of inertia are negligible, the effects of any one of the other asymmetries are small, but the effects on vehicle performance of some combinations of the asymmetries may be catastrophic. 2. Unacceptable vehicle ;erfcrmance is likely If either resonance (resulting In large lateral loadi) or zero roll rate (resulting in large impact dispersion) occurs during reentry. In either case the asymmetries must cause both a trim angle of attack and a roll torque, Therefore, a center of gravity offset combined with either an asymmetric pitching (or yawing) moment or with nonzero products of inertia or a pure rolling moment asymmetry combined with either of these asymmetries is apt to result in unacceptable performance. 3. The conditions under which adverse performance is likely tv occur are dependent upon many variables so that each vehicle must be analyzed separately. Nevertheless, the following trends may be stated. The probability of encountering difficulties resulting from combined asymmetries is increased for missiles, o o 'o o o Having low stability margin Having resonance at an altitude near maximum dynamic pressure Having low normal force per degree angle of attack arid/or a low pitch damping moment Flying shallow trajectories Having slender shapes
o
4.
""
J"
UNIVKRGI
4
In the derivation of the equations given in Section 4. 4, it was assumed that the body was symmetrical. We consider next the effects on vehicle motion of asymmetry in mass and external shape. Asymmetries in mass and external shape result from design and manufacturing tolerances or from inflight phenomena and may cause one or more of the follow
Ing anomalies:4
Aerodynamics
1. 2. 3. A change in normal force characteristics A change In pitching moment characteristics A lateral shift in center of pressure A pure roll torque. Mass 5. 6. 7. A lateral shift in center of gravity Nonzero products of inertia Unequal pitch and yaw moments of inertia.
4.
i .
C.g.
e. g.
and
L
" A,.
= . .
167
. . .: i : : : , :? :: . , I
. .
..
4. 5.
Unequal pitch and yaw moments of inertia Pure roll torque, I.. A
The asymmetrLes have negligible effect on all portions of the trajectory except for the reentry phase. To account for the asymmetries, the equations given in Section 4. 4 must be modified to the following (see the sketch on page 170):
0b1X 0 1  qr(IZ  1y) +Xy(4l  pr)l+lxzlf + pq +l+Izlq"  r2 B4,1)
i!
2 
I(+I
. pq)
(4, 52)
(4. 53) (4.54) (4.55) (4.58)
.1
lI) 
2 + 1y(p
q2) + 1 +1
(0
"m"F AX+MqF Z
n
+tn
FyAX + Mrr + FY
rc. g.
+n
4.5.1
that only an asymmetric and no) Assuming exists, Eq. (4,.51) reduces to: X = Mpp,
(4. 57)
Therefore the roll rate decreases slightly under the influence of viscous damping, M . The trim equations F+ F qq+m AX + Mr +no =x+ pr(IyIX) r X (4.58) (459)
.,
r
168
Y
,I
AI
5
I
UNIVtNSITY RYLAND
Consider first the trim conditions for p * 0. The pitch damping terms are negligible for this case, and the trim
equations become: m
z
0
510)
AX n0 a0
(4. 511)
in terms of a and j (Eqs. (4,413)
(m . MCmoqSd and no
CnqSd).
o
0 Nd AX Cno eN
.(4.
512)
[(C)2
+e(Cn)'I
" AX
_C___
cN
and the plane of
by the equation:
N
tf
AX
(4.514)
cot
,
Fy
0 o
Cn
(4.515)
',.I
..
,....
...
""
,
..
..
...
......
,;,,.
1,149 JOHNn
APPLIKI
HOPKINS UNIVVG
?ITY
PHYSICS LABORATORY
V
I~I
+Za'.F~
Fo
o "0 011to
900
i3 s
and as s
14
Sattack,
The angle 0o also defines the plane of the asym,!I metry. _That is, for zero roll rate the total angle of jyol is in a plane containing the asymmetry. For example, if the asymmetry results from a flat spot on Sa 0 or 000o + is 180 (depending 0 the body, flat located upon the fore and at a circumferential location aftis Location ofthe the asymmetry).
di
tpn
te,
The effect of ro roll rates to modfy the magnitude and to cause the total lateral force to be rotated out of the plane of the asymmetry. These effects are given by the followdng equations (Ref. 46) based on theory given tn Refs. 4o and 48. 4 of
170
.1
.
2
I
+ D2x2
1/2
2a
,I
0
1(1
(4., 516)
FV
(4.520)
1Y_
Pr
A
(4.521)
From Eq. (4. 51G) it is apparent that the aniplification 1'ictor increases from 1 at X = 0 to a maximurn value
[D (
2211,2
at X
at X S. Thus, at X greater than about 'V2(depending upon the value of D), the trim angle of attack Is attenuated by the effects of roll rate. Typical values of D are of the order of 0 to 0. 2 so that Amax occurs near X w I and lID. The condition of = 1Is called the pitchAmax yawroll, or simply the roll resonance condition. Since 171
iMT
O JCM ,J M
UN W IVIIVT PMNSwp~
0. 1, the resonance condition Is very close to the SIX/ly condition P/WA n 1. Note that near resonance the pitch dam~ping term, D, plays n very important roll in limiting 0.
The value of D is always positive for conditions (static and dynamic stability) for which IEqs. (4. 516) and (4. 517) are valid, Therefore, from Eq. (4, 517) for positive roll rates the plane of ;. for a rolling vehicle is rotated counterclockwise (A0 is negative) from 0o. At resonance, the plane is rotated 900 from oo at very high values of X, the plane is rotated 1800 froin 0o.
+j
F
F
ISNEGATIVFE 134
zI
..
For example, assume that an angle of attack is produced by a flap at the aft end of the body as shown in the sketch on page 173.
I
'172
1l
I<
T,K JMHNI HOPKINS
APPLIE"
1 I
Y y .(
W!(
Z
along the Z axis. As the roll rate increases in a positive direction, F rotates counterclockwise. At resonance F " FR t Fo/D and is directed along the Y axis. As the roll rate increases further, F continues to rotate counterclockwLse and decreases in magnitude. In the limit (very high roll rates) F  0 and is directed along the +Z axis. To further illustrate the behavior of Eqs. (4. 516) and (4. 517), Fig. 413 shows plots of A and 60 as functions of X, for D = 0 and 0,.2. Note that neither A fnor A0 changes very rapidly for X from 0 to about 0. 5. Practically all the change in A0 occurs from X I 0. 8 to 1. 4,
.
F0
.
For
'
is attenuated.
The effect of damping is small except near X = I where the effect on both A and A@ is large. As the reentry body descends through the atmosphere, WA increases, and unless the body is given a very high spin rate the resonance condition will occur at least once during reentry. If o were zero, the resonance condition
173I ,
fH't
JOHNGHOPMINSUJNIV9NNITY
1I6VIN SPRIP40 MARYL.AND
would have no effect on the trajectory. If the only risymmetry were the asymmetric moments (n,, and in ), the effect of resonance would normally be minor since p is cessentially constant (changes by viscous damping only) and WA would pass through the resonance condition too rapidly to pi'oduce significant changes in the trajectory. The a would inecrase somewhat over the ry value, resuiting in correspornding increases In body lift and drag, but the time duration would be short, Therefore, the effect of an asymmetric moment, by itself, has negligible effuct on the trajectory for reasonable values of mo andA no* 4.5~. 2 Offset Center of Gravity The effect on trim conditions of the center of gravity offset (again neglecthig M1) .'or p 0 is given by:
F .
Co
g,
(4.522)
Fy
Y X
c~.*(4.
AX
523)
%C
C N
AX Y *A
(4. 525)
2112
174 
IFATI
For zero roll rate, o, is in a plane containing the center of gravity, and F acts through the center uf pressure directed away from the center of gravity as shown by on". However, for a positive roll rate the force vector F is rotated counterclockwise from F as explained previously, Therefore, F results in a negative roll torque that tends to decrease the roll rate, A similar argument for a negative roll rate shows that the center of gravity offset results in a positive roll torque, again tending to drive the roll rate toward zero, Therefore, the effect of the center of gravity offset alone behaves as a roll damping effect and, in fact, must be accounted for when attempting to extract C, from expvtrnmental roll rate decay data. For
p

175
.......
.......................................................
"
lii
i:
i'i l
l i
ll
i I
....
Jil
' '
.
Ullle'4d
S&q.AILA
reasonable values of the center of gravity offset, the effect on roll rate is small except near resonance where the roll moment arm. is maximum (see sketch above), 4. 5. 3 Combined Asymmetry
(Coand r)
Both the center of gravity offset and asymmetric momem' result in minor effects on the trajectory when considered separately. hlowever, this is not the case when the two types of awymmetry are combined. Using the described procedures, the trim angles of sttack at zero roll rate for this case are given by the equations:
Am ci
0
zA
S=   :..AX
(4. 5 2: 7)
'
Cy
CA 0o d no
A
cN
(4. 5B)
=:
(4. 529)
The direction of F
V
cot 00
00
o ' " :.
. .. ..
 
(4. 530)
M Ad
Z c. g. 76

The proper quadrant for 00 is again determined as indicated in Section 4.5,1. An Important variable for this combined asymmetry is the angle, 0, between the plane of the center of gravity and the plane of Y.
I
F
c.g. 0
where
(4.531)
otI"c.g. = 00 to
c. 9. g.. if Y is + and Z
(4. 532)
9100
is +
= 90* to
1800
if
if if
is
and z
is +
It must be assumed that the center of gravity offset and moment asymmetries are completely independent variables so that 0) nay have any value. For purposes of illustrating the significance of 00 it is convenient to consider "in plane" asymmetries (00 = 00 or 1800) and "out of plane" asymmetries (00 900 or 270*). For the in plane asymmetry, if 00 = 1800. the effect is the same as that
177
.
w :
APPLIK9
PHYSICS LABORATORY
described previously for the center of gravity offset. If 00 = 01, the magnitude of the roll rate will increase indefinitely (5 is always positive) if the initial roll rate is positive and will always increase negatively if the initial roll rate is negative. This case is shown in the sketch.
Y.1
F (pl
At p 0, Fo passes through the center of gravity. If the roll rate is initially positivrt F must be counterclockwise from F. as shown, so U... a positive roll torque develops driving p to a larger positive value. Since F must lie between and 00  1 800, the roll torque Is always positive. Similarly, if the initial roll rate is negative, the roll torque produced by the asymmetries is always negative. For out of plane asymmetries, the case of 0o = 900 is sketched for an initial positive roll rate. Let the asymmetric force be located initially at F (counter clockwise from F since p is positive). The roll torque is negative and p decreases. For decreasing p, X decreases (assuming wA is constant) and F rotates clockwise. The roll rate becomes zero when F  Fo and then the vehicle begins to spin up in the negative direction. This process continues
"178 
,I
'I
fl[c
JOMNiSm HopiIp
UNIVRII!y
RI
Z
0F
until F rotatos
900 from F0 . This point is an eqailibrium roll rate position, since at higher negative roll rates the roll torque becomes positive, driving F to a less negative roll rate. This equilibrium roll position also corresponds to the resonance condition, For 002700 the same phenomenon occurs except that the roll torque causes spinup to resonance at positive roll rates and therefore does not spin through zero roll rate.
The potential problems resulting from these combined asymmetries are: 1. *2. Excessive spin rates resulting in structural or payload failures Prolonged periods of roll resonance resulting in high lateral loads and high drag (structural failure and excessive impact dispersion) 3. Spin through zero roll rate resulting in large impact errors (see Section 5. 4. 2).
179
h.!
:.."
_
,II
Both In plane and out of plane components of the
asymmetry can have aisastrous results onvehicle performance. However, the in plane asymmetries are less likely to cause difficulty because sfgrnficant roll torques are developed only at conditions near resonance. Also, this type of asymmetry cannot produce the problem of spin through zero. 4. 5. 4 Effect of Comb.nod Asymmetries on Roll Rate All the potential problems resulting from a combination of asymmetric moment and center of gravity offT set listed in Section 4,5.3 exist because of changes in roll rate during reentry. In this section we will consider the roll behavior in more detail, If Iy  Iy and if the products of inertia are zero, the roll acceleration (14q, (4, 5.1))becomes: Mp+ Z F+Y
Y 0g,
Z C. g.
(4.533)
Ix
Neglecting the roll damping and pure roll torque terms and expressing the numerator in terms of F, r, and 0: Fr
'
sin0
IX
(4,.34)
Since F
_qSo'r sin
0
IxI
.
~(4.535)
Sil
?MI JNt
.O3PN1
UNIV94iITY
where
=G
sin 0
0
A sin 0
(4.537)
'I
(4,53[3 )
The G term contains the effects of roll rate and 0o; the JnOmax term contains the body geometry, environmental, and basic asymmetry terms. Note that b maxis the positive roll acceleration that would exist for zero roll rate if the plane of a were normal to the center of gravity plane (that is, 0o  270"). Since 0 = 00 + A0, Eq. (4,537)
may be written..
(4. 539)
Substituting for tan A$i and A cos AOp from Eqs, (4. 517) and (4. 518), Eq. (4.539) becomes: DX cos 00 (1 \2) sin 0 G222o +D2 X2 (l X) X 2 +D
(4.540)
0 as the roll equilibrium condition, For We will define this condition either Omax or G must be zero. For a 0 reentry body in the sensible atmosphere ( 0), is zero only if co or r is zero. Generally, neither ao nor r is zero and the roll rate of a reentry body having combined asymmetries will vary until G = 0. Defining Xe as the value of X at G = 0, from Eq. (4. 540), >e=e or Xe must satisfy the equation:
tan 00
DAe DX 2

'
(4.541)
D cot 0
o 2
F
xe
cot 00
or
(I +
cot 0)0
(4.544)
Equation (4. 542) is plotted as a function of D cot 00 In Fig. 414. The roll equilibrium condition, or unstable; that is, if the roll rate is X changes slightly from Xe' If the roll at X tends to drive X back to Xe, the Xe, may be stable changed slightly, torque developed equilibrium condi
tion is stable; if the roll torque developed tends to drive X. away from Xe, the condition is unstable. The Xe condLtion is stable if (dG/dX)X (dG/d%)>e is positive. given by:
3~
2
'4
2 +A sini0[D, 1)24X
D oo0 ( 2 D 1
182
______________________
~~IlL.VII
(4. 546)
Z
(4.547)
2  x4 J 2X[(D 2  1)+ 2x
For in plane asymmetries the condition for stability reduces to: Cos
0oll
+(2
D)X 2  3
<
(4. 548)
Since Xe = 0, the condition is stable for 00 : 1800 and unstable for 00 = 00.
For out of plane asymmetries, the condition for stability reduces to:
sin 0X
(D
1)+ x2
o0
(4.549)
Since X. = 1, the condition is unstable for Xe = +1 and 0o = 90* or Xe. l and 00 = 2700; the condition is stable u00, for Xe = +1 and Oo = 2700 or Xe = 1 and Oo = Another characteristic of interest is the maximum values of G. The term maximum is used to mean either a positive or negative maximum. Maximum values occur at values of X obtained by setting the numerator of Eq. (4. 545) to zero giving the expression: tan 0 = F(X, D)
,
(4. 550)
For 0Oo = 0 or 1 800, neglecting terms smaller than D2 , Eq. (4. 550) may be solved for values of X and Gmax giving the result:
183
, ?. ..
_.
.J
Xk
2 D ( N  n)
D!
(4.,551)
Gmax
(0 + ) cos 0 D i
G t  for 00 WCO
(4. 552)
'A
Therefore, at X f+ I
k~m i
max
G max p G
XA+ 1
xLfor max D
0o
180o
XPdl1
1 f ~1,B0"10
For Oo = 900 or 2700, Eq. (4, 550) may be solved for X, giving the following equation: V1 From Eq. (4.540): +V(4.553)
max
sin D(2 D)
(4554)
The value of G is the same for k. The sign in Eq. (4. 554) corresponds to the D term In Eq. (4. 553), Therefore for 0 = 900, Gm wx 1/2D at X u Y1=, and Gmax  IY2D at X = . For 00 270*, Gmax el1/2D at X= V= and Umax l/2D at X V =T'. Note that, for a given Fo and r (i, e., Omax), the potential . '11 torque for an in plane asymmetry Is approximately twice that of the out of plane asymmetry.
184 
14 1
SI.',
Sll
'I
?"t JOHNI
MOPKINI
UNIVIWtIITV
7 ..
For the general value of 00, the maximum valuns of G may be obtained from a plot of F(X, D), Since ~F(X,, D)  F(X, D), It is necessary to plot only the +X. values of the function, The maximum values of G for +X are obtained at MF(, D) a tan 0o; the values of X are obtained at F(X, I))  tan Oo. A sumnmary of the characteristics o(f G is given in the table for in plane and out of plane asymnmetries. 01
QM
Poi/v
01 .
.
Umak
2 D i
, MAX
11 4.5
270. x 0 MA
maxx
"rn
Negativye
I/il
I *1+.
1/2D]
t4TW"
.lID
ID t./I2D1
I,
0
Unatabil'
*t
Unatable at +1 nStable at 1
(3
tabllt l S
St ble
tAbl
It 4+1
saltable
at "I
Typical plots of G versus X for D n 0. 2 are shown for 00  0*, 450, and 900 in Fig, 415, The values of G at a given X for 00 = 1 800 and 2700 are the negatives 00 O0 and 90% respectively. Note that of the values for for in plane asymmetries the roll torque from combined asymmetries is small except for a narrow band of X in the vicinity of 1 (i. e., near resonance conditions). However, for out of plane asymmetries the roll torque may be appreciable for all values of X between about 1. 5 to 1. 5 except at X 1. The function F(X, D) is shown for D = 0. 2 in Fig. 416. Using Eq. (4. 550) as the condition for Gmax, note that, except for small values of 0o (less than 00  45), X 1. Furall values of Gmax tend to be nearX = 0orX thermore the magnitude of Gm x tends to be small when it occurs at values of X other Ean 1, For example, the maxima for 00 450 or F(X, D) = I are shown in Fig,
185
1
'I 1"..,.,,.,,'.. .
416 to occur at X" 0,12, 0.'74, and 1,06. From Fig. 415 the corresponding values of Gmax are 0. 7, 0, 9, and 4.1, respectively, 4. 5. 5 Roll Resonance One of the potential problems resulting from asymnietries is prolonged periods of roll resonance. During this period the angle of attack increases, resulting in large lateral acceleration and high drag, If the resonance occurs during the heating portion of the trajectory, the problem may be amplified byjdistortion resulting from asymmetric heating since at a one ray along the body remains on the windward side of the body and is heated more than a corresponding ray on the leeward side, In this section we will consider conditions under which prolonged periods of resonance may occur. The term 'lockin" will be used to describe a roll history in which Pr tends to follow WA so that X f constant. That is, 1 . It should be noted that this does not. necesS/1/A sarily imply that resonance (X  :hl) exists. However, asymmetries are usually small enough that roll amplificration is required to cause significant changes in roll rate and the necessary amplification can occur only for conditions at or near resonance. Typical values of roll rate at reentry are I to 2 Hz, and the static margin typically Le 3% to 476 of body length. For these conditions, the peak value of WA is much greater than PE so that X = I will occur at least once during reentry. If the range is short enough or the reentry angle shallow enough, the resonance condition may occur a second time, as shown in the sketch on page 187.
4,
iI
. . .
I
ThE J@MNUlM@PWINUluNiygRU'ITr
FIRSTN
RESONANCE
SECOND
RESONANCE
WA
ii
"1
A vehicle having no roll torque would exhibit a reduced roll rate (Pr) trace shown by 1.. The roll rate would
pass through "first' and "second' resonance without incident. However, if asymmetrie~s exist, the roll rate may
Curve 2 is typical lockin on WA a shown by 2 through 5.
~of the case of large asymmetries that exist early in the reentry phase. in this case lockin may occur at "'first" 1n the more usual case, asymmetries develop resonance. I'' as a result of the heating environment so that "spinup" "first" (cases 3 and 4) or "spindown" (case 5) occurs after S. resonance. For cases 3 and 4 the angle 00 is between 1800 and 3600 so that "spin up" occurs; for case 5, 0 is  between  . 00 and 180* so that "spin down" occurs. Cases ?and 4 indiCate that p may lockin to WA (case 4) or may cut through the WA curve (case 3),
..
~.
...
I..
The roll lockin conditions are the same whether the vehicle has positive or negative roll rate at reentry, 187K
THE
To simplify the presentaiion, only positive roll rate will be considered In the remainder of this discussion. If the reduced roll acceleration (Or), produced by asymmetries, is equal to or greater than WA, lockin may occur. In equation form, lockin may occur when:
._
(4.555)
ly
Using Eqs. (4.437) and (4. 318), it may be shown that 4A is given by the equation (neglecting variations in temperature with altitude):
FC
WAVE2
,K
Using Eqs. (4. 38), (4. 536), and (4. 538), f~Is given by
the equation: G VE M
2
m sinCN " r
K '
2 IxCA
(4.557)
od o
I
yd l
siny
Iytr
121
/ ,,
(4.558)
"IN
Equation (4. 558) may be simplified further by noting that, for a given altitude (or given value of K), the minimum
L,
188
. . .
rum ~~ ~
.
~.,.nI~lIWflln ~~~~~.
...........
asymmetry that may result in lockin is given by Eq. ax. If trim conditions are realized (4,558) for G " is given approximately by G 10)5then 4.;, Section (see (see Eqs. (4. 552) and (4.5 4j): Gmax =k/D where k has the following values: (4.559)
0
900
900
90
I
12
1 +2
0. 5
1 2
l
1920 80
2700
1  2
2
0.5
UsingEq. (4. 520) and the definitions of F. and Mq, an expression for D Is:
D
where
k1
Cmq]
(4. 560)
k, =
0 (
'y
I
ymd Y
Substituting Eqs. (4. 559) and (4. 560) into (4. 558), and noting from previous discussions that Gmax occurs near the resonance condition for in plane or out of plane asymmetries: r R (4.5,61)
18 9
I
LL
H
7',
FT7
?NU JW4,,NI M,OWKIF4 uNIV IItNBiI
where
F1 =
/I
 c
(4.562)
and KR is the value of K at the re,,onance condition, The general behavior of Eq. (4.561) is shown in the sketch,
4MAXIMUM
ad 2
ItMUST ItS
*)
"W
k M UL ST 6 E
';
'
LOCKIN POSSIBLE
NOT POBSLK
NO
POSIL
NO
KOS OLE
REENTRY
dP
 19 0 
TIM9
HOPllUMS JOW4NS
UNIV&PIT'Y
Following reentry (K
(0
vided k> 0 and lockin will continue at least to K = I unless dynamic effects (roll raWte nd n oscillations) or roll damp3reak ing effects (sec Section 4. 5. 6) cause "break out."
out will occur at Klj = 1 unless 00 is such that and therefore G change sign near X 1. Therefore, from
Fig. 415, break out will occur at KR = I for in plane asymmetry (00 = 00 or 1800) but may not occur for out of Break out would plane asymmetries (00 = 0or 2700). occur, however, even for the out of plane asymmetry at K 3 . To be assured of freedom from lock in at all
conditic'ns: or (ry )
whichever is the more restrictive. The subscript 2 refers to the second resonance condition. Neglecting a change in roll rate during reentry, it Since the slope may be shown that 11  K 21I I  K I. of the negative the of the ( r/d curve for K = 0 to 1 is more the slope for K > l, the (O r/d)1 limit is always resonance the fI'om Therefore, restrictive condition. standpoint, it is desirable to select a low value of PE so that first resonance occurs at as high an altitude as possible. The maximum asymmetry allowable is obtained by evaluating Eq. (4. 561) at K = 0 so that to avoid lock in:
r
<
(4.563)
In Eq. (4. 561), the asymmetries are given in The ar':ymmetric moment coefficient terms of e and r. is a better parameter for some purposes since the moment coefficient is a direct function of the asymmetry whereas Co depends also upon CN, and vehicle stability (AX). The components of a are shown in the sket'ch on page 192.
.4.
".
,Ik
"
"'
'
"
THR JOHNI
HOPKINS UNIVRRSITV
IlL.VIIII~
~~
0PhaMI.N
SIN
CASE FOR 00 900IS SHOWN
01
zI
The component of a resulting from the center of gravity offset is always directed away from the center of gravity, whereas the component resulting from asymmetric moments may be in any plane. The components of a are (see Eqs. (4. 527), (4. 528), and (4. 529)): Cr CA dx
CN
o
~c.g.
(4, 564)
Nd
ax
=o
Cm
N d
(4. 565)
where
=

Cm
n2
(4.566)
192
, . ,.
., .
I
?149 JOMN% HOPKING UNIVIRIYTV
For
toward the center oi OM must be directed Then>&T. gravity and nOM >Ca . o00o
r
SA(4.567)
Cd
For 0o =180: + C.A.
m
a . AX
r Ad
For o
=900 or 270*:
"21
ag <<y OM
[c
Usua ll
C CN mo
9.
(4.s 60) v
"
(iK
F A< F
2
(4. 571)
"193
+i)
.
where F
r, A
CC
r m~ 0 d
1C
CA
m
0
r
fcr
'0 0
0
0
dor
,
00
A
Cm o
(4.572)
C 2
1
0
for
900 or 2700.
F 2 =C N F I*
(4. 573)
The Iimitirng asymmetry values may be obtained for any in plane or out of plane asymmetry using Eq. (4, 571) with the appropriate value of k given in the
X . 1, G is no longer negligible and, if the f'r produced at pointo 6 (0 = 00, 90% or 2700) is equal to or greater than (;A, lockin may oc.ur at some value of X between
a and b where G is just large enough to satisfy Eq.
(4. 555).
approached, (A decreases, becoming zero at K = 1, and then it becomes negative. As a result p must reverse 00 nor 00 900 sign if lockin continues. Neither the 0 00 asymmetry case can produce the characteristics required
194
 . +.. . +, 
. u q nrol
impM
00.90
0 / , .L
tt,
TO PITCH FMIOUlINCYv'."
IsD
I
i
"*for
1195
greater than that given by point c. For the 00 =180" case, lockin cannot occur since k (or Gmax) is < 0, The vehicle would pass through first resonance with a slight reduction in roll rate, Both and eotend to cause a rapid transit (both effects tend to reduceAX) through the reso$I nance condition. As WA increases, X decreases to some low value and p remains nearly constant. However as Wdecreases at K > 1, A. increases. Lockin at this time (second resonance) requires a negative and may occur between points b and c if at c is large enough. For Oo 900, the shape of the G curve is such that lockin
could occur at second resonance between b and c. However, if the torque at c is large enough to cause lockin
.I
CA
0
(4.574) 
"<
r < dF .2 T
t'.575)
From Eqs. (4. 562), (4. 573), and (4. 575), note that the allowable asymmetry is governed by the following parameters:
I'
II
_A.I:
196
Cm r to avoid lockin
Parameter Vehicle stability margin Resonance altitude Planet of asylmmetriesa Aerodynamic coefficients Type of trajectory A tmosphere Vehicle slenderness Mass distribution factor Vehicie size
CN and C N m q
sin y
H
Low CN and Cm
Low y (shallow reentry angles)
High H
IX Iy n C m I,, /md d
Low Y,'
a"~r Cn
Since there are so many variables, it is not possible to present generalized curves of allowable asymmetry. However several observations are noteworthy. Consider a family of bodies differing only by a scale factor. Values Of CN, ~Cmq. IX /Iy, and Iy/md 2 will be approximately the7 same for all bodies, If it is further specified that all bodies have the same stability margin (same AX/d), then for a given atmosphere and trajectory, Eqs. (4. 562), (4. 573), and (4, 575) show that the allowable CM r is proportional to d , Therefore, small bodies may be particularly sensitive to mass and aerodynamic asymmetries. Note also that for a given allowable F 2700. For 00 .00, the minimum occurs at:
d2 CA
AA
TMI9 JOHNS
HOPIINS UNIVZISITV
Co<2
For Oo
Ad 0
GF 1  K,
(4.576)
<m
rm /
(4.577)
c CE
lr0
< 2 29 CA
2 (1
KR)].
(4,578)
Therefore for a fixed value of FA the minimum value for "mCo for an out of plane asymmetry is about 15% greater than that for an in plane asymmetry.
and I
p
(4.579)
max
198
~Ii
"where
G'
G +G 1 +G G2 (4. 580)
The amplification factor G i defined by Eq. (4. 537). factor G, is defined by the equation: Mpp max
V
The
'd
CN
Vd
c AdX
1cN C01
0 "Iy
[AXK sinT]/
GI 1 2HO A k I'r
CY
)'=a).
12'1
, (4.582)
The value of a is always negative. Therefore if the G curve is drawn, the GI = G + G, (assuming G2 : 0) values are the ordinates measured from a X1 axis which is rotated counterclockwise through an angle, c, where: tan( =a 199r,, I
(4.583)
I,
0
I
Similarly: C G2
0o 0
0oa
C
(4. 584)
Xmax
od
If G 0, thL values of G' r `, + 2 are the ordinates of the G curve measured from a X axis that is shifted by
C
an amount C.
I'
200
..
..
'
* ....*?.,
..
*." "'*~'
" :"
d'
:'"
"
'"
"
jIAPPLIED
tWNIVRRSITY
Suppose an in plane asymmetry exists and ao and r/d are such that an ampli.fication factor, G a GR (see Eq. (4. 558A, could result in a lockin condition at first resonance if CI 2 0, Lockin may then occur at point A in the sketch.p
G~ G
C I
e2
201
I
I
"TP41
JOHN*
MOPKINB UNIVIRpITY
then is obtained by drawing the line BC as shown, giving a lockin condition at B. As resonance continues, K increases and therefore C increases, If ( exceeds the value shown by (2, the torque required to sustain resonance is unavailable and break out occurs.
tan
C *
Using Eqs. (4, 558) and (4. 560), and noting that R max G G "aG R Gmpx  GR' it may be shown that roll
max damping will cause break out when the following inequality is satisfied:
2 (H/d)CAklv
r
k
KR
CA
md2
((.585)
J
" Cmq
sin (,y)
Roll damping tends to decrease the possibility of lockin at first resonance and, if lockin occurs, tends to increase the possibility of break out. The opposite effects occur at second resonance. When CIO exists (but not C) lockin cannot exist
unless
GR
1
T
Using Eqs. (4. 558) and (4. 560), lockin cannot occur at
first resonance if Cp has a sign opposite to that of the initial roll rate and a magntude given by the following inequality:
I, ,.
202

~
0.
t r
1 od
R R
a1 K
,K
(4,.58(3)
where
CR
> 0
. N k = 1 Cnnq
CAX k
kC
sin ()
(4.587)
a2
1
yI~
(4 588),
If the sign of CIO is the same as that of the initial, roll rate, the effect of CIO is to increase the probability of lockin; if CI and PE have opposite signs, the effect is to decrease the probability of lockin. 4. 5. 7 Spin Through Zero Roll Rate
In the previous sections, the problem of roll resonance was considered. A second problem resulting from variations of roll rate during reentry occurs when the roll rate becomes equal to or passes through zero. When p = 0 and an asymmetry, go, exists, the lateral acceleration resulting from is fixed in space and the missile moves i0 off course in the direction of Foe The impact dispersion produced may be much larger than that required for satipfactory performance. In this section we will consider the conditions for which spin through zero may occur. Referring to curve 5 in the first sketch in Section spin down through zero roll rate may occur when
4. 5. 5,
203
I!
is in a ThereThe enough
an out of plane component of asymmetry eAlpts and direction to reduce the magnitude of the roll rate. fore, one condition for p = 0 is that 180 " 00 > 00. second condition is that the magnitude of is la: ge to result in p = 0 prior to imprict.
For the asymrnmetries of interesi in this problem, the vehicle usually passes through first resonance with little change In roll 'ate so that mrnt of the change in roll rate occurs at ), noar zero so that roll amplification is negligible, The roll acceleration, then, is given by Eq.
(4. 5w35):
.(4.
589)
.x
1 dV .g dt
CA S W
Therefore:
" d" qS A
"
(4. 590)
CN
d0
r sino
0n m
dV dV
(4.591)
A .
i
If p and V ace assumed to be the only dependent varla.. blE.s, Eq, (4. 591) may be integrated to give: CN P P
A p
rr sina 0
roVE( '
).
(4. 592)
204 
Ii
APL ED PHYS:C.S
LABORATJI',
up4..vfRr,
~
=
~CNo

i
d V0E K/2

4p 1 C
(4. 5f)3)
A X
(IN/rod() at I<
1(
If
LAP
> P
<SL'
to impact. %rhealtitude (or K) at which zero roll rate occurs may be obtained from Eq. (4. 593):
PE A m2
Pc
IX
e =1 /2 
N(4.594)
(Y
sin~ ,
1 (4.595)
or
K =2 in
I CN
PEA md 2 ,
Ci
 r VTE
sin
L N
ad
The minimuin asymmetries which may cause zero roll rate occur for 0) = 900. For this case, Eq. (4. 595) may be written in terms of the moment coefficient using Ea. (4. 570):
In
I 
PECA C~ m2 P m2
(4. 596)
d
o d Mn
L2
m
0
If spin through zero is to be avoided for the entire trajectory, Eqs. (4. 595) and (4. 596) indicate that the following inequalities must be satisfied:

205
LJNIVE.SIIr'
E Vrd 0o 1 p
A
2
md
CN
( PS.L.)
(4.59 7)
s in_ ,
or In terms of C
FA
oA *
1 (CA2 C (C m 0
r2 d)2
(4.5a98)
S~zero
ff
206
Tht
II
A=cos "
A[
0o
r/d
and (go r/d)rnin is the value of &0 r/d given by Eq. (4. 597). For this assumption, the probability of encountering spin through zero varies from 0 at io r/d = (o r/d)min to 50% at ;'o r/d :i (0. r/d)min' 4. 5. B Effect on Products of Inertia The products of inertia are defined by the following equations: IXY I I mXY MMm X Z (4.599) (4.5100) (4.5101)
i ii Yz =EmYZ
The IyZ term is negligible for most reentry bodies and is omitted from this analysis. If Ixy and IXZ are not zero, the principal axes pass through the center of gravity but are rotated through an angle 8 measured clockwise from the body centerline. For small asymmetries, the rotation angles are (for Iy = IXy
Iy
(4.5102) x
SI
"" y
I(4.5103) X
*1
.
207
t;W
(4.5104)
where 8 about the is a rotation about the Z Y axis, 0 but axis and 6 is a rotation
IXy.
/1
Assume the vehicle is symmetrical except for two small masses, m 1 , located at (X 1 , Y 1 ) and (X 1 , Yi) so that S=2mlX 1 Y 1 , If the missile is now rotated about the X axis, the centrifugal force acting on each m 1 produces a yawing moment about the Z axis. Returning to the case where Ixy 0, I # 0, and the mass asymmetry forces are much greaterthan the aerodynamic forces (X . cc), and retaining only the larger asymmetry terms (p is moderately large but 1Xy, 1Xz, q, r, , t, and tarc assumed to be sufficiently small that the square or product of two small terms may be neglected) of Eqs. (4. 51), (4. 52), and (4. 53), the equations of motion become: 208
S.
1
TI JOiNG HOPKINS UNIVIENIIITY
SIl.VUR
0 2 IXZ =rwg  p
(4. 5105)
(4.5106)
2 IXy y
g P (4.5107)
where w
= P11

j.
S=
tf.
(4.5108)
Using Eqs. (4. 5107) and (4. 5108): +Wg 2 q =gP 2 Xy IY (4.5109)
j
(4.5111) 4.5112)
(4. 5112)[
,t
XZ
209'r )1
q p
+ r( p
'x
+C
p I IXY
In other words,
the telemetry traces for a body having nonzero IXy and IXZ show a q which oscillates about the value q
the~ teeer
rpY
I
y
and an r
p I
The con
CI
tP PyI
X
(4.5113)
I C2 :qi P.Iy
(4.5114)
tan 0
(4. 5115)
I
210
.. 1
I
T""r JOMNS HoPSINi UNIVgRh1?V APPLJED PHYSICS LABORATORY 0P0.NG MARN6AND *1Lw.V
The period of the rotation about the space cone is given by Eq. (4. 425). Since the body centerline rolls around the princLpal axis, the angle of attack oacillat('.; ith a relatively long period (%r) and a peaktopeak amrlcude of 20 with a superimposed short period (l/p) oscillation with peaktopeak amplitude of 26. The constant shift in q and/or r resulting from nonzero products of inertia may be interpreted in terms of angle of attack using the following equations (see Ref. 49),
"1
*q
)I
*
'p
1y
XY i Xz
X
(4,5116)
(4.5117)
aerodynarnic forces are no longer negligible, assuming that tyo and A. may exist as well as ab and b the total
angle of attack is given by the equation (see Ref. 49):
A (
x2
2 + (e5 + +2)2]
(4.511 )
where A is defined by Eq. (4. 516). trim force is defined by the equation:
0
where
00+ A
(4.5119)
"tan
0
(4.5120)
o0.
ii
if
and
tan 211
(4. 5121)
~Ii
ifa CO
0, and
N
Ktani(.
or
2~+j,
tan'
a
(4.5122)
Ai
if either
=tan'
and 13 or ei
D 2
and 0.
The proper quadrant for 00 may be obtained from the signs of the numerator and denominator in Eq. (4. 5120). For 0o  go = 0, the trim angle of attack is just that resulting from products of inertia. Let
%=
2(4.5123)
is given by: (4. 51 24)
* Iil
212 
*1
The amplification factor ts zero nt ), 0, Ias the same amplification as that for aerodynamic trim at X = I (namely, A), and approaches unity as X. . Therefore the products of inertin may be important at X near or grenter than unity and, Ln combination with the center of giavity offset, could result in resonance lockin. The effect may either augment or diminish the effect of other asyrninetrius, depending tpon the signs and magnitudes (see discussion below) of Ixy and IX . However the product of inertia asymmetry in combination with the centcr of gravity offset cannot result In spin through zero since the roll torque rt p = 0 is zero, Consider a vehicle having a positive roll rate at reentry and a combined inplane asymmetry consisting of a center of gravity offset along the +Y axis and a positive product of inertia IXy. At reentry (X w) a positive 1Xy results in a positive g (Eq. (4. 5116))or a force F,, along the Y axis. Since F rotates clockwise as X decreases, F is always negative and the roll acceleration resulting from F is always positive, and therefore lockin at first resonance is a possibility.
FtF
Xf ,
/:
!!/

/!
Z
23 
Solving Eqs, (4. 51) and (4. 52) simultaneously and retaining just the b contributions resulting ftom aerodynamic forces:
I
x
+ XY m
I I and y m using Eqs, (4, 54) and (4. 55):
Substituting for
Fy Z c.g.
F AX I Z XY
Ix
y
Note that fbresulting from the second term on the right side of the equation is negative since Fz is always negative. had been selected Fz would Similarly, if a negative Iy always be positive. Ther efore this term results in a negaand, for positive roll tive P regardless of the sign of Iy rates at reentry, tends to prohibit first resonance lockin but tends to cause second resonance lockin, In a manner similar to that used in deriving the criterion for lockin for an aerodynamic asymmetry and a center of gravity offs'ct (Eq. (4, 561)), the criterion for lockin with a product of inertia and a center of grRvity offset is: r
2
IX
'
This inequality is identical to the inequality for an aerodynamic awymnietry and a center of gravity offset (Eq. (4. 561)) only for the particular case of a neutrally stable vehicle (AX/d  0), which is not of practical interest. For nny staticall stabie vehicle (positive AX/d), j r/d is :tJwaj;i greater than eo r/d for a given F 1 and KR. Solving for rY 8 at the boundary (the above equation set to zero) using the binomial theorem:
214
1
___
__
APPLIED PIYSICS
ASAB0RATnRPe
(~":~) ~
'411(F
)2j
(4.J5125)
where
11
F4 1=1
KRi
8 a min
where an and
66 max
8
the negative and positive signs of the radical term, respectively. The general requirements for lockin for this type of" combined asymmetry are as shown in the sketch for a particular value of K
:*
215
"I!!
LA6
lASYMIPTOTE
LOCKINLOCUS OF
r ,., (Fo
a6
AT 7i4
ASYMPTOTE
0
3
( m
,n
KR I
F3
"'
Note that reet r id)t decre y sses from t7 at reentry (KR. = 0) to 0 at maximutii dynniic pressure (KR and then increases to/"' I1K . i. <t impact.
.', ;'. [
Current
reentry vehicles are designed sulch thant first resonance occurs at Klit 0, so that I Klt 1. Assurniing p does. not change during reentry, it may be shown that II  K~t >1 at second resonance. Therefore lockin is not possible at first or second resonance for any value of '6 if the vehicle can be constructed so that:
216
1
"1t
______
V
JOH6NS it,"*SU O
____7_P_
24T
SL~~i
LUtL~
fo (
VI' a Id F3thi's
inequal~ity b~ecomes
genstcI/& tin~~
X Y
q
si
1/2
*'
tedt
be inldependent of size.
the allowable center of gravity offset varies as d32and [sin (/)1 1/2 Therefore, small bodies flying shallowangle trajectories are more susceptible to lockin than are large bodies flying steepangle trajectories. 4. 5. 9 Unequal Pitch and Yaw Moment of Inertia For a body whose only asymmetry is unequal pitch and yaw moment of inertia, the equations of motion (neglecting sniall terms) become: QY II Z)qr (4, ~12 6)
= mn Ni
pr(IO
(4,5127)I I )

=n pq(I Y
4 518
1)qr term in Eq. The significance of the (l sketch. The body in the Iz > IY) shown (for 51 26) is (4 snmall eq~ual masses for two except symmetry has mass from the center (AY) distant equally axiE Y on the located of gravity. At some instant, the bod4y has t. transverse rate of rotation shown by the vector T At this time the
THE
JONNIN
OPKINS UNIV9I.mI'Tv
rn
II
r
t
body rotates about an axis coincident with T and the centrifu al force acting on mI produces a couple 2mlAY qr = (ty  IZ)qr about the X axis. The roll torque is negative when IT lies in quadrants I and III and positive whenT lies in quLadrants II and IV. At very high altitudes (X m), the terms m and n in Eqs. (4. 5127) and (4. 5128) are negligible and the equations for q and r are those given by Eqs. (4.418) and (4.419): LT Cos , t (4, 5129) (4. 5130)
WT sin w t
At t; 0, C T lies along the positive Y axis and, for positive p, rotates counterclockwise with constant amplitude the asymand frequency as t increases. Therefore, occur in roll rate metric mass and the rotation of WT result in a roll rate oscillation of cycles Two oscillation. for each revolution of WT about the Z axis. The equation for the roll rate history is obtained by substituting Eqs. (4. 5129) and (4. 5130) into (4. 5126) and integrating to give:
 .
, ....
2..
,'1
I
t~ 01M HOPRIP46 UNIVEM6ITV
"_"_
 sin wait
(4,5131)
where
gl Pt1
('
Therefore, the roll rate trace in the exoatmosphere oscillates as a squared sine function with a peaktopeak amplitude of:
2p P\Iy
2 2 Iy  I(Z
Ix/lx
wgi.
When the vehi.cle descends to altitudes where aerodynamic forces are no longer negligible, the ZT develops a pattern of motion that depends upon the initial conditions and the vehicle's aerodynamic characteristics, Unless the rotation of ZT about the Z axis becomes very slow, a net change in the magnitude of p is unlikely and the effect on the reentry trajectory is rnegligible. Under some conditions it has been found (Ref. 410) that the direction of rotation T about the Z axis actually changed directions during reentry. During the period when the rotation changed sign, a significant reduction in roll rate occurred. The possibility of encountering difficulty as a resuit of unequal Iy and IZ appears to be remote, and this type of asymmetry is not considered further in this report.
219
4.5.10 Applicability of Trim Characteristics at Resonance In preceding subsections, it was assumed that trim conditions are nttained at first resonance, However the resonance condition is approached very rapidly, and large ciLnnges in trim occur. Unless the missile can respond very quickly to the changing trim connuitions, the missile
will pass through resonance without attaining large ampli.
SWA.
fications in the angle of attack. The weathercock frequency, is an indication of the ability of the missile to respond to a change in trim. IfWA is high the missile response time is short; if WA is low the missile response time is long, Since
*1
P
WAy
AX4S
x1
WA is proportional to 't Therefore, no matter how much static stability (AX) the missile has, the response time is always long at very high altitudes. If resonance occurs at a sufficiently high altitude, trim conditions are not attained and the loads and lockin criteria based on trim equations are not valid, The conditions for which trim theory is valid have not been examined in the literature. In order to provide some insight into this problem, a brief study was made of the lockin criteria using a sixdegreeoffreedom trajectory simulation. Trajectories were computed for three bodies haying the following mass and aerodynamic characteristics:
II
ILI
220
I
TiUB JONUS HdOPKINSl UNIVIKIITY
*,
Characteristic C A CN
Body A 0. 104 2
Body B 0.104 0. 88
BodyC
0. 104 2
'4
1 0,114
1 0.261 3
31.1 3.5 30
d
1
Body A is the body used for all examples in this report, and additional characteristics for this body are given in Table 21. Bodies B and C have the same value of F. (s:ee Eq, (4. 562)) as has A, Therefore, the lockin criterion (based on trim conditions) for a combined asymmetry consisting of a center of gravity offset and an aerodynamic trim (see Eq, (4, 561 )) is the same for all three bodies. However, for a given altitude, WA is higher for B and C than that for A by a factor of about 3, and therefore, for resonance at high altitudes, B and C tre more likely to attain trim conditions than A. The trajectory initial conditions used in this study are VE = 20 000 ft/sec, YE = "300, and hE = 400 000 feet. An in plane combined asymmetry was used for the first set of calculations, The asymmetry consisted of a center of gravity offset (r/d  0.001) along the +Y axis and a positive yaw moment coefficient (C,,o). The altitude for first resonance was varied from about 40 000 to 140 000 feet by changing the initial roll rate. For each initiRl roll rate, C, was increased until lockin occurred, Out of plane asymmetries were considered using the same
'
221
...
_41111111.. .. . ..
...
...
. ,.
uenter of gr.vity offset with positive and negative values of pitch moment coefficient (Cm ). Values of eo are r'Lvctly reantecd to Cn nnd C mo(see Eqs. (4, 567), (4. 5(i8), rindt (4. 569)). ThreU aspu1:tS of trim theory are important in the
2.
S;o;or/;) is l/D.
3. At X : 1 the plane of the angle of attack has rotated 900 from its (defined by A6) value at reentry (K  0).
The results from the sixdegreeoffreedom trajectories for body A with PE =l0000/sec (corresponding to first resonance at 89 000 feet)show that none of the three assumptions was valid for this resonance altitude. The magnitude and plane of W lagged the trim values and the peak amplification was much less than l/D. The sketch shows a comparison of the K and A time histories as given by trim and simulation results. The trim 9 history reached a peak value at X  I where lockin was assumed to occur and then decreased with increasing time, since D increased with time. The trim value of IA0I decreased from 1800 to 900 at X = 1 and remained at 900 as time increased, The simulation results show that a built up to a maximum value but lockin did not occur, Following amax, the missile began a coning motion with no large excursions in roll rate. The peak ev was about onethird of the trim value at the initial first resonance altitude; the peak a occurred about 1. 5 seconds
222
"TRIMTHEORYTWITH
LOCKIN AT
,, 
SIMULATION RESULTS
.90,
'
\I II
Ii\
I
00
\\\
w00
00
later, or about 15 000 feet lower in altitude than that corresponding to first resonance altitude; the plane of rotated through an angle of 90 about 0. 7 second later, or about 7000 feet lower in altitude than that corresponding to first resonance altitude. The angle of attack amplification effect results in larger values of the asymmetries required for lockin than those given by trim theory. Under some conditions the lag effects would be similar to lowering the first resonance altitude and, thus, asymmetries required for lockin based on trim theory would be too high. However, for altitudes where first resonance usually occurs, the amplification effect is predominant and the required asymmetries for lockin based on trim theory are too low.
223
The condition required for lockin is given by Eq. (4. 558). If trim conditions are attained, then Eq, (4. 55B)
may be simplified to Eq. (4. 561). Both equations are plotted in Fig, 4 17 for body A for an in plane (k = 1 ) and an out of plane (k = / 2) asymmetry. Note that the assumption that trim conditions exist implies very lnrge vahues of the roll amplification factor G at thu higher altitudes. Also shown are the at r/d values required for iock.In obtained from simulation results, At the lower altitudes "trim theory provided the correct criterion for lockin for inplane asymmetry; at the higher altitudes the values of r/d required for lockin obtained from simulation results were someowhat higher than those given by trim theory, In fact, the lockin boundary is very nearly.a constant G contour. The results also indicate that the ao r/d value required for lockin for the out of plane asymmetry is about twice that required for the in plane asymmetry as predicted by trim theory. Similar trajectories for bodies having in plane asymmetries were calculated for bodies B and C for an altitude of 136 000 feet only. 'These results (Figs. 418 and 419) show that bodies B and C achieve an effective roll amplification of about 18 compared to a value of about 12 for body A. Although B and C have the same response, C was much closer to a trim condition than was R. Note that for an altitude of 13U 000 teet the trim condition for C corresponds to G 27, whereas trim for B corres pends to G , 60. These results indicate that w.A is one factor which affects thu attainment of trim but it is not the only factor, and much work is required in order to define conditions for which trim theory is applicable, Trajectories were also computed for body A for a of a product of consisting asymmetry inplane combined inertia andsa center of gravity offset. In this case Ci8. Was held constant at 0. (6360 and r was varied until lockin occurred. The altitude for first resonance was 89 000 feet. The results show that lockin occurred at a value of
'
224
S.
8 r/d a 2.6 x 5 If the asymmetry had been an aerodynamic asymmetry and a center of gravity offset, the simulation value of a r/d would have been 1, 2 x 105 (see Fig. 417). This ratio of about 2 to 1 between the two types of combined asymmetries is nearly the ratio obtained using trim Eqs. (4, 561) and (4. 5125). Therefore, although trim was not attained for either set of asymmetries, the relative behavior in terms of lockin criteria was properly prre( o.'d using trim equations, Until a.Iltional data become available regarding the applicability of the trim equations, the following comments appear to be valid: I. Trim equations pro,.ide a valid criterion for the lockin condition provided the roll amplification factor implied by trim is not too high, 2. At resonance altitudes where trim conditions imply very high values of G, trim is not attained, and trim theory predicts asymmetry boundaries that are too low. 3. Values of go r/d required for lockin at very high altitudes tend to follow a constant G contour rather than that defined by the trim equation. 4. The limits of G for which trim conditions are For attained are probably a function of many variables.
the bodies, trajectory, and initial reentry conditions used in this study, values of G greater than 10 to 20 were not attained. Until more work is done on this problem, trim theory may be used to estimate the asymmetries required for lockIn. However, the results for conditions corresponding to values of C greater than about 20 should be considered as questionable. 5. For an out of plane asymmetry, the required o r/d for lockin is larger than that for the in plane asymmetry as predicted by trim theory. Also the relative magnitude of combined asymmetry, a and r, and combined asymmetry, o and r, are proptrly defined by trim equations. 225 
,I
SILVI
00OMPLN
'
!I
W..
II
22
t5
"i,
tHI
sIpol'" MAtit&ND
'
.2
.3
.4
3
.2 U
13
4
,
COT )I
Fig, 4.14
I2

22B
THI
t I
qr4I
00
229
or
4ow
0I
I
to
ILI
23
2 30 0I

I ~'
10 4
Gr13
OUT.OF.PLANI
10I
1001
10.61

Fig. 417
231.2~4
TH& j: ON
PII '
NI,
"
10r
02
I
4.
,
EQ 46SB

EQ 4.B61 RESULTSI
  SIMUJLATION
40
60
s0
100
120
140I
Fig. 418
232
4
INPLANE
40 60
eo
10
iao
40
'
C CRIERIA BODY5
233T
SIUATO
E'4.I6
Ik
r
'm
,.Ot4NN
MIOING
UNIVUIrTv
IL
2351
w,......
23
I
T~4UJOINWI M@PgMINSl w.IvmsIYElv APPLIED PHYSICS LABORATORY
SILVRR SPRtIN MARYLAND
SUMMARY
Examples illustrating the material presented in Section 4 are presented in this subsection. The following specific proibIems are considercr: Example 1  Minimum energy trajectory for nonrotating earth Example 2  Ballistic trajectory for rotating earth Example 3  Body motion (oscillation) characteristlcs in the exoatmosphere Example 4  Histories of reentry velocity and related factors Example 5  Body loads as affected by a combined asymmetry (center of gravity offset and a distorted body) Example 6  Roll lockin aN affected by the combined asymmetry of Exclmple 5 Example 7  Roll lockin as affected by a cornbined asymmetry  center of gravity.A offset and nonzero product of inertia
'I
237
*1i
4, 6, 1
Example I
Compute the characteristics of a minimum energy trajectory (MET) for a nonrotating earth for a launch from
0L 0, A 1 = 0* to a target at Tho range is(Eqs. (4. 2It = 3U10 nmi,
OI
The range angle is (Eq. (4, 218)) 280  2 8e The azimuth angle at launch is (Eq. (4, 220)) 3540. OLI For an MET, the initial flight path angle is (Eq. (4. 110)) o * 450

150 = 30.
From Eq. (4.111) o 0,815 and, assuming the initial altitude to be sea level, the initial velocity is V
0
ha
The velocity at peak altitude is (Eq. (4.124)) Va  15 500 ft/sec, or about 73% of the initial velocity. The time to impact is (Eq. (4.117)) t secondrc. 4.6,2 Example 2 1480
For a rotating earth, determine the characteristics of a trajectory fired with initial conditions V,, Y0 , and OLI (relative to earth) computed in Example 1. The inertial quantities V,
v', and
at launch are
239 
I.
t iL
tan
tan
=
. . 1
1 5 20( l)
, 5 4 =n 1 1 4 2
OL
VL
39). ILI
1520(1)1 O.036
22 000 ft/sec
21 200
0.0363
sinX sin y
y
0. 48 2
=2.
Therefore the effect of earth rotation is to increase the initial inertial velocity by about 3. 6%, to lower the flight path angle by 1. 10, and to increase the azimuth angle by 3. 70, These changes would have opposite signs if the vehicle had been fired in a westerly direction. Using VL and V1 for V. and 70, the equations given in Section 4. 1 apply. Howeve'r, note that the trajectory is not an MET since Eq. (4.111) is not satisfied. For the given VL a minimum energy trajectory would haveVL = 28. 1 so that the trajectory being considered is slightly high conmpared to an'MET,
240
..
.'.
The corresponding range, r0 ( 2 0 ) =4020 nml,is the arc length traveled in inertial space (or the range on a nonrotating sphericnI eiarth for initial conditions VLV, ~L The impact time is (Eqs. (4, 15) and (4, 116))
t=,1660 seconds.
The impact latitude is unaffected by earth rotation
and is
450.
49,. 40.
370
nmi
A summary of the results is given below: Non. otating Quantit Flight time (seconds) Impact latitude (degrees) Impact longitude (degrees) Impact range (nmi)
Earth
rotation is to increase the range by 170 nrni and to increase the flight time by 180 seconds.
46.3EapeThe typical reentry body, having mass characterinthe exoatmosphere, At separation, the body Is spun about its longitudinal axis at a rate of 3600/sec and,
241
?NIs JOQ4m
uNIVIRU11tY WOPIKINBl
s.
onds
The somicone angle of the space cone is (Eq, The body rates measured by, onboard 590, (4. 423)) 0 rate gyros have the following characteristics (assuming a symmetrical vehicle): 1. The roll rate is constant and equal to 3600/nec.
2, The pitch and yaw rates very sinusoidally about zero with peak values of 600/sec. The pitch and yaw rates are 900 out of phase. The period shown on the trtices is (Eq. (4, 424)): TB
1,11 seconds,
If the body were spin stabilIzed so that the total angular momentum vector It were along the flight path_ direction at reentry, during the early phase of reentry a Iowever, the pitch would be constant and equal to 51). and yaw (inertial) components of a would oscillate with a halfamrlitude of 590 and n period of 5. 12 seconds.
i,,
I
tI4 .JOHNS WOPSI~i4 UNIVSPIYI'
'I
4. 6. 4 Ixample 4
I
I
Determine the effect of the atmosphere on the motion of the reentry body using the following reentry condi
tions,.
hE V, ,00 000 fleit .
2(00 00 ft/se
.300
PE 0/sec and ',h01'/sec (6. 2H rad/sec)
I
, J,
0 given in Table 31 and Atmosphere brd Use the 1962 Stnd To anle o ptb.m. in T.Able, given the mass Lind aerodynamicv characteristics
conical motion rs problem. not in a The 0, the thes vehicle 3Io was SincemTb as described in previous angle of atta ck tv remains constant and in the original plane a pitch nd yaw angles of attack remain constant) (Inertial bw until terodynamic forces begin to affect th e motion byI as the beginning Using tu0.d99 th in dtcb y cause of then 01.a decay, for . PE 3600/sec:  (Eq. . 64.5)) 5 21 x 23. f0e7
1(
Po
From Tfable 31, this density ratio corresponds to an altitude between 300 000 and 350 000 feet. A more complete table shows the altitude to be h = 332 000 feet. If the Initital roll rate were zcro, Eq. (4. 4.41 ) could be used to obtain the altitude at which the decay begins.:
' '
. 5(2)(0o 4) (21TY
2 96x1 :3 e t
)xo
fe
'
I
_243_

I.
~APP'LIED PHYSICS
I.AEpRATCnRY
Using thle charancteri1stics of' thu 1 !)6i2 Standarid Atmos~phere, this value or 11 corresponids to nn nititudu of' about
30/ soc spin 350 000 feet. rkherufrc, the uffuct of' tih:o~ ra~te Was to de'lay th0 W deOCay by abOUt 18i 000 fQLt.
VIW C( )1IflptCd 0811Wj Thu ;; lu(Ca Ili qtotr iL'LA 11. \' Eqs. (4. 444) and (4. 453), In UidUl' to evaluateW thcsuL
Those characterlst&H, obtnined using Eq~s. (4, 36) and (4, 323) and the atmosphore defined In Table 31 , are listed in Table 41. Prvom Eq. (4. 445) and Table 21, tile danmping constant is: required.
About 60% o f C:i comnes from tilt CNr termi ind 40% from
the Cm,
q
term.
was computed using Eq. (4. 45`), 111d flt)u VUSUlS arc shown in F ig. 420, Also shown is tile ease of the nonlrolling missile ond the ctise of n rolling miissilec with ing term is negligible for altitudes above about 150 000 feet. Vor nltitudos bolow about (GO 000 feet, the W history for tht. rolling ca so (with damp ing) is noarHy thle sarnc 11s that for the nonrolling case. 1However, nt altitudes above 60 000 feet, thle angle of attack at a given altitude for the rolling body is significantly !ifgher than thnt. for the nonrolling body. This fact could he inmp ortaint for' some( nmissilos in which the punk lateral lond occurs at these altItudes. For the body studied, thle peak lateral lond resultK 4 (7.6 + 0,75) .0 09)0

From Table 41, this value of K correspondH to ain alti tude of about BBi 000 feet, The altitude for manximuni is the same whether or not the missile is rolling, Howfor thle over, as ra result of the ev histories, g1

~.tt
244
I
?MR JOHI MOPKNINNUNINIrTv
I
rolling missile is about 45% grenter than that for n nonrolling vehicle (Eqs, (4, 460) and (4. 463)). lor = 101, 1E g~n for a rolling vehi.cle is 2. 75g compared to 1, 9g for a nonrolling vehicle, The reentry velocity, Mach number, and Reynolds number arc important environelental factors that affect hody Ion Its and tciperrturu, The time hi story of these characteristics estimated using Eqs, (4, 35), (4, 37), and (4, 310) and the atmospheric characteristics given in Table 31 are listed in Table 41, The Mach ni'! r history shows that the vehikle Is in the hypersonic rrlime (M t (6) down to an altitude of about 15 000 feut aid is supersonic fronm 15 000 feet to impact. The Reynolds number history shows that transition on the body woulcj be expected at an altitude of about 110 000 feet (RL ft 10 ) with lamknar flow at higher altitudes ind turbulent flow at lower altitudes. The heating rate to the conical surface of the body would increase by a factor of 2 to 3 when the flow becomes turbulent, and therefore asymmetries that result from heating would be expected to increase rapidly at the transition altitude. The aerodynamic frequency data show that roll resonance
x~~ 1i)
1 35 000 feet. It should be noted that the vehicle is approaching a second resonance condition at tImpact, and any asymmetries that exist near altitudes of 135 000 feet and sea level tare likely to be t.implified. The computed values of M/h. show that free molecular flow regime begins at altitudes well above 400 000 feet (M/Reo 10), transition flow exists down to an altitude of about 250 000 feet (M/'"e `, 0, 1), slip flow exists at altitudes from 250 000 to 140 000 feet (M/V'.' 0, 01), and continuum flow exists at altitudes fromn 140 000 feet to son level, The time from 400 000 feet to impact is about 49 seconds compared with about 39 seconds if the itmosphere were absent. Therefore, tne atniosphere int vcased the flight time by about 10 seconds,
:1
245
N1.
MOHN.
HOPKIN.
UNIVEISIT
Although not shown in the tahic, severni other importnnt consi deratiovnts may 130 obtninud from the equa tions given In Section 4, 3. Using the rntio qs/qtlonx
0,1 as Lin Indication of eondlitiOnH wherLe heattinog is s ignif'icant, the n1titudes ore computed (PI i . (4,. 311) anl(. (4. 310()) to be about 190 000 feet for the ueginning of thw hunting peril ld 1 1l)Olt t111 1 3 000 f[U t forV If . end of heatoing, With the n11;dm11u. hujntink Vntte ot M 1 nhltitudtI or :Ibout H11 000 fQuet, Thi, h0logitu lt innl I .et'nt ( f s (4, and (4, :314)) Is maximum at an altitude of Tbout :5 000 feet where m 54g. The inflection points in the gN versus time curve (Eq. (4.,319)) occur nit altitudes of 5B 000 antid 19 000 feet. The results obtained in this problem nie usunim. marized In Vi'g, 421. Although the nltttude,bankds shown . .
tory diharcteristics, the results are typienal of' high J3 re': en~try bodies. Note thant practtectlly all of thu quantitttem of interest occur at altitudes holew about 1,110 000 f'eet and
Both laminar
and turbulent flow conditions exist In this altitude region, SaUl three Mach number regimes occur (a'though only two occurred for this trajectory), and both oscillatory and trim types of body motion may be important.
4. ,5 .
kI
hxn m l c
Suppose the reentry vehicle has a center of gravity offset from the body centerline of 0, 042 inch. lF'urthermore, assume that as a result of exontmospheric binst effects the body is distorted so that the body centerline is an arc of a circle and that the tangent to the centerline changes by 0. 250 from base to nose tip, Use the saime
246
I
I 1"~HkJCOHNI MOOK~iN5 UNIV9111ITV
II
body is 0. 002. We will assume that the lateral location of the center of pressure and center of gravity are unaffected by the bent body. Flrom Eq. (4. 565) the trim angle of attack resulting from CM iq: 0 0.0,02 57 3 0 0M 2x 0, 114 x57,0,5,
S18
The trim angle resulting from the center of gravity offset (4,.564)): (Eq.
e o
Therefore, the total trim angle Wo is very nearly the angle resulting from the bent body and we will use 0 = 0, 50. The equation for gl is: CN S
4L
For the moment assume that no amplification in a. oceur.c, so that A = 1. For this case, the largest possible vtlue 'mx cI: I From Eq. (4.313) occurs at q of at P/Po
2116 2110
(.)
= 0.
q max
54 400 fl/ift
247
,1,
Excessive values of gL will not be obtnined unless values of A in excess of 2 are obtained, Amplifications of this magnitude will be encountered if large excursions in roll rate occur. The possibilitios that must be consiclered arc: 1.
2.
3. 4.
Spin up with lockin at some intermediate altitude, Spin down with lockin at a negative voli rat,.
I
,
Assuming that lockin Is attained, trom .1Eq, 1/iD and D may be evalur':,d usihig El, (4. 516) A (4. 560). Substituting these equations and th c utivtlo ' 1, excecssiv, for q (Eq. (4, 38)) into the equation 1. loads will occur when the following inc.untilty is sattified, 57,3 x 20 (k0o; 2 C gd, A" d
C NIN N7 E
where co is expr.isad in Jectory p ar:metter ., feet:
1 '
,y
,d.oree...
LI.k,
Substituting
'orthe' U 1.
1..ll ,
S0. 04
I
leIF;
This Inecqualit, is plotted in Fig. 122. It i..i u,; ,. .', 0, 50 is large, (h:i,;U!1 to ,'sult ill eXC";:V' I.i.... that io 1c at some altitudes, Tho fir':8t od second rescmntc( .IIi are the altitu,les for whiclh WA =,
PI
Ise..
248
M 111 ""NA
"APPLiED PHYSICS
7M` JUNIN
MOPtINS
WNtV9RttTY
LABORAT;)RY
[':
Values of wA were computed using Eq. (4. 323) and the results nre shown in Fig. 423. First resonance occurs at a. altitude of about 89 000 feet, and second resonance occurs at an altitude of ahout 4000 feet. Therefore roll lockin at fir'st resonance would result in excessive loads, hut lackin it second resonance would nlOt. i['Depending upon the value of Oo, the roll rate may
istics. However, the magnitude of Ap in this case will be underestimated slightly since the amplification in o and the change in the angle 0 are neglected. Using Eq. (4. 593) and assuming values of 900 and 2700 (which give the largest values of Ap) vL,iues of Pr were computed and are shown in Fig. 423. From this consideration it is found that the asymmetries are not large enough to result in roll reversal. Therefore, resonance at negative roll rates may be eliminated as a possible condition that would result in excessive loads. However, if the vehicle does not lock in at first resonance but spins up, resonance may occur at an altitude of about 13 000 feet. From Fig. 422, resonance at this altitude would result In excessive loads. Therefore, the problem is reduced to an evaluation of whether or not lockin will occur at altitudes of 89 000 and 1 3 000 feet. The criterion for lockin is (Eq. (4. 561)):
0rd
(I KR)
Using the data given in Tables 21 and 31 and the trajectory char cteristics, the value of F 1 (Eq. (4. 562)) is 5.4 x 10 for h = 89 000 feet (H = 21 700 feet) and is 4.7 x 106 for h = 13 000 feet (71 25 000 feet). Using r/d 0. 042/12(3. 5) 0. 001 and expressing 0 o in degrees:
249
S.
I
mml mmi m il l ! l
...
ml~~ l ii m
. . .. . . .. . ., ,.
d i iI[ii INi I III ! I'a]
THI
S(1 0. 3 1
Kl
f'o h1 W) 000 foet
(i
> 0, 27
K H)
k
for 11 1 :R000 1'cct,
Using the data of T:ible 41 a plot of K versus altitude shows that KR 0, 077 at 8,9 000 feet nnd K 0 2, 5B at 13 000 feet, Therefore, the criteria for lockin are:

S
ok
0, 286
S0.427
j
I
.IZ
The value of k depends upon the value of 00 and, for lockin to occur, k must be poitLve for I<R <1 and negative for Ki > 1. For this prohleL we must assume that 0o may have any value. Therefore, we must consider the following possibilities: Resonant Altitude 69 000 feet 0o 00 !}00 2701 13 000 feet 2700 k 1 0, 5 0, 5 0, 5 o 0, 286 0, 572 2 0, 572 0. 854 ,
The other twu values of 00 having negative values for k for the `13 000 foot case are ruled out since the 00 = 2700 value is required to provito spin up during reentry (see Eq. (4, 5;)3)). O'f the four possibilities onl~y the==0o :00 case with resonance at 89 000 feet satisfies the & crite rion, In the preceding discussion w' considered only inplane and outofplane conditions, ].or intermediate values
I I
r
250
3
,. .. . ', , '. ! " " .;', . . .. .. ' p .. '.
S.
IJ
of 00, io may be divided into inplane and outofplane components. Any value of 00 for which the component of ;o in the 00 = 01 direction is z 0. 286 will result in excessive values of gL The possibilities are shown in Fig. 424. Any value of' 0 defined by points from A to B 0 (shorter arc length) will result in excessive values of gLSince any vnlue of 0o is equally likely to occur, the probability of encountering excessive 's: iL . 283 2 cos"1 0 360
F
f
0.3
4. 6. 6
Example 6
Sr :!igiven '["
For the trajectory used in Example 4, estimate as a function of PE the maximum allowable &0 combined with m 0. 042) inch so that neither lockLn at first resonance nor spin through zero occurs. In view of the discussion In Section 4. 5. 10, assume that the criterion for lockin is the trim equation (Eq. (4. 561)) or G  20, whichever gives the larger value of &o. The critical conditions to be investigated are the 0o S 00 case for lockin and the 00  90" case for spin through zero. For the lockin case, assume that H = constant 21 700 feet. Substituting the mass, aerodynamic, and trajectory terms into Eq. (4. 558), the criterion for lockin becomes:
2.65 (1 KR)
If trim conditions are attained, the criterion for lockin is (see Example 5): P4 0. 31 (1 KR)251
'I
='"" ................................................ 4_ _''''..... '...... "'.
Tlti JOHNSHOPK(INSUrRII
APPLI1EDl PHS SS
LAIBORATORY
20 in the first eq~tifl Sett IngG0 the ssme solution to ii0 2.3 0. 3
th.
wo
qaifSg
A~
or
70 000 feet (Sce corresponding to an altituido of about > lower than 70 000 feet (or K 4 ~an Table 4i). At ,11titudies is higher Rltitiide8 at used, 0. 18~3) the trim equation aqllatiofl is used with 0 20. 70 000 feet tile noe1trity correponling.ThLiF~or each xesonant altitudo the equation:~ the frorn tial roll raite ME'Y bno Computed WA
Ix
c Thle valuesore where WA Is given in Table 41. i. functio Fi. qurdfrlckin are shown a~s 425. from asymmfeThecangein roll rate resulting Spin through nd r/d is given by Eq. (4, 5133). &p is negative tries; 0 Fi if reentry zero roll rate will occur during condition restr'ictive most Using thle &p : pE PE In units. and expressinlg for (00 9011 and K rK )andres~pectively, critevion the of degrees and dog sec, spin through zero roll. rate is: 41 spin through zeru r'oll rate Tihe values ("f i; required for are shown in Pig. 425.
4and
This problem illustrates some of the tradeoffs which must be considered by the designer in selecting a roll rate for the reentry body. If the roll rate is very low, first resonance occurs at a very high altitude, where large values of 00 may be tolerated without encountering lockin. However, vehicles with low values of PE are susceptible to Lin through zero roll rate. On the other hand, for moderately high values of PE, large values of 0 may be tolerated without spinning through zero roll rate, but oo must be very low if lockin is to be avoided, For this particular problem the best roll rate Is PE = 7600/sec where the missile could tolerate up to 0. 60 in &o without encountering either adverse condition, A second possibility would be to select a value of PE much greater than 22400/sec. In this case spin through zero roll rate would be unlikely, and lockin is avoided by using a roll rate that Is much greater than the maximum aerodynamic frequency. This technique is sometimes referred to as "overspin." 4.6. 7 Example 7
i1 '
Using the trajectory of Example 4 with pE = 10000/sec, first resonance occurs at an altitude of 89 000 feet (KR  0. 077). Determine the values of 8 and r for which lockin at first resonance may occur. Assume that the trim equation (Eq. (4. 5125)) is valid. Compare the results with o and r given by Eq. (4.561). 5.4 x 106 (H = 21 700 feet). From Eq. (4. 562) F 1 Using the mass and aerodynamic characteristics &iven in For 8 in . 26 x10". 4. 87 and F 4 Table 21, F 3 degree units, Eq. (4. 5125) becomes:
rI
ei 279~.l 12,5
The ea boundary is shown in Fig. 426. At a given value to must be greater thanl6min but less than of r, Lockin cannot occur for any value of 253If r < 0, 06 inch. i8
.max.
I,,
.j..
o and
0
be
The oo boundary is also shown. It Is noted that &0 Lhas no Omax boundary, and the W. boundary is less than *6min Single points on each boundary obtained from sixdegreeoffreedom simulation resulas (see Section 4, 5. 10) are also shown. The (G. r) point was obtained by holding r constant and increasing no until lockin occurred. The (;6, r) point was obtained by holding q constant and varyIng r until lockin occurred. Note that the real boundaries are somewhat higher than the boundaries based on the trim equations as discussed in Section 4.5.10. For other values ofEpE, the ea boundary must remain within the x and &m asymptotes shown in Fig. 426, and rmin is shifted right or left depending upon the value of KR. Using data given in Table 41 and the relationship between pE and WA at resonance,
WA
E,
X
of P' For r
=
lockin could exist only for values of PE that result in resonance very near to the altitude of maximum dynamic pressure (35 000 feet). For current reentry vehicles, values of pE are selected such that first resonance occurs at altitudes above 100 000 feet. In this altitude range rmin is insensitive to PEI and the &6 boundary shown in Fig. 426 is applicable to first resonance lockin for all values of pr!,
254
[
u n n mI I
I
TH
JOINNI OPImNI, UNIVRRSIT
BLS1U SPPNIIi MAImAN,
less than about 10000/sec, For Px 1000*/sec considered In this problem, second resonance occurs at an altitude of about 4000 feet (n'nglecting a change in roll rate during reentry) wlev'e rmin is nearly a factor of 2 greater than rmin at fir.7t ;"4sonance.
4 I
1
,1
I
255!
255
!
* *.I ~ ;
.4ui~q .. . ..
REFERENCES
41
L. S. Glover, "Approximate 1ceEntry Velocity and Heating Equations, " J. Spacecraft and Rockets, January 1966, D. T. Greenwood, "Flight Mechanics of Space and ReEntry Vehicles, " Michigan University Engineering Summer Conferences, 1964, H. J. Allen, Motion of a Ballistic Missile Angularly Misaligned with the Flight Path upon Entering the Atmosphere and Its Effect upon Aerfdynamic HeatIng, Aerodynamic Loads, and Miss Distance, NACA TN404B, 1965. R. B, Powell and R. L. Smith, "Dynamics of Spinning feEntry Bodies, " AIAA Preprint No. 64470, 1964, H. I. Leon, "Angle of Attack Convergence of a Spinning Missile Descending Through the Atmosphere," J. Aerospace Sei. , August 1958. L. S. Glover, Analytical Expressions for the Effect on Roll Rate of Mass and Aerodynamic Asymmetries for BallisticTypeBodies, APL/ JHU TG 560, March 1964. J. D. Nicolaldes, On the Free Flight Motion of Missiles Having Slight ConfigurationaL sy'mmetries, Ballistic Research Laboratories, Aberdeen Proving Ground, Report 858, June 1953. R. L. Nelson, Measurement of Aerodynamic Characteriatics of ReEntry Corifigurations in Free Flight at Hypersonic and Near Orbital Speeds, Lockheed Missiles and Space Division, TR 6906137, July 1961.

42
43
44
45
46
47
48
257
.ij
REFERENCES (cont'd)
49
A. E. Hodapp, Jr. and E. L. Clark, Jr., "The Effects of Products of Inertia on the Ru .1 3Behavior of Ballistic ReEntry Vehicles," AIAA Paper No, 70204, January 1970, L. S, Glover, "Effects on Roll Rate of Mass and Aerodynamic Asymmetries for BallistIc ReEntry Bodies, " J. Spacecraft and Rockets, April 1965, K. A. Ehricke, Principles of Guided Missile Design, Space Flight, Vol. Ii, "Dynamc D Van Nostrand Company, Inc., 1962,
410
411
2I
$'
 258
Li
.... ,, . . .! . ' , .1
I"
ji1
/
/
I:

l fl7/""uDM'
ROLLING,
DAMPED
it
00
'2
280
320
360
F~g. 4.20
CHARACTERISTICS a DECAY
I/
259
ii,
I
I
>.
400
\L)
:360
.I 320
....
INiTIAL ,i DECAY
2(10
280

S240
200
S HEATING BEGINS
120
,
\)OUNUARY TRANSITION
LAYEi
0go U.
RATE Iq51I
40
PE A
,IT, AN DWA
10
20
30
40
60
Fi,4.21
OIGNIFICANT EVENT,
260
~ ~~:?
.( ,
It
26
SONVNOSIW 0140039
3Y
26
I_
T111 JOHNG OPKINI NIE1VIRS1
mmi
!I

LOCI'.IN CRITEHIA
0.6
III
0.4 A
001
II
0 04OB
F0.
42.
C DEG REES
0.21
o.B
II
0.4
i0.6
TWE JOIHN
200 000 4
1
'tl 2S2
L~flt(IN RES1ULTS
oSPIN
150000
400
800
1200
1600
p000
2,400
2800
PE (degrees/seood)
Fig.4.25
F,
 264
.i1
.................
,,,
x olo
I.
~z
LLLL
IA.g
00
""
10
Q. ,
..
0,
pnoeel~p) 9V NV0
SI
265
"
0,12
ALTITUDE F4USONANCE S L
"
0,08

! pIRST RESONANCE
o o
goo~ ro
oo
004 0,04
00 2009080070
400
800
1200
P dlgrS.I~tSi~Ord1
1600
2000
2400
Fig. 427
266
Ilk
LJNIVINEIlY
[4J
in
mi
C,
04 N
c N dl
I,
C NC
i N
0,
00
0
tM~
to
t.
Ni
In
C 00
~
nD
C M
0000
~
mi
C"
C;
14J
cn0
0
CI
in~ 0
.
00
t
to
C"',~
267
4A.
r,.
. DISPERSION
[
Preceding page blank
3!'
?rek Jo,l
4OPKINS
WNIVII:".ty
Symbol A AR CA CD c. g. CL CL .C o
Definition axial force component of A in dowrnvange direction axial force coefficient = A/qS drag coefficient = D/qS centerofgravity location lift coefficient = L/qS dCL/d& asymmetric rolling moment o/qSd coefficient A moment coefficient about body Y 0 (see Section 2.4) axis at
Cm 2 nC
"Cm 0
*Cm
0
I
=
CN
SdC N/d&
C no c. p. D d E e FW F moment coefficient about body Z axis at " 0 (see Section 2, 4) centerofpressure location drag, aerodynamic force parallel to V vehicle reference length factor defined by Eq. (521) Napierian base weighting factor for wind Weighting factor for density variation
L
' . . .. e.
L
271
..
i
. . . ..
. wl
Symbol
g 9gY g g Y gZ
Definition
acceleration of gravity
Ty_.cul Units
I't/sec 2
1 2
+ gJ
21
1/2
g
H h 15 I
IX
moment oa
slugit.
index number P K L
sin Y
lift, aerodynamic force normal to V asymmetric rolling moment pounds ftlb
pounds
0l'
ambient pressure vehicle roll rate
or3
lb/ft2 rad/see lb/ft nml 2 1716 ft 2 01
 oi*
2 dynamic pressure, pV /2
/0ec
feet
y0
272
Symbol
r S S. L.
Definition
rodial c. g. offset reference area sea level
Typical Units
inches ft2
IT
ambient temperature
vehile mV velocity
O
ft/sec it/see ft/se . ft / se c ;
i}I..iVR w VA/M
drift velocity resulting from winds windspeed velocity of the air relative to the missile
W X o S
vehicle weight
distance parallel to the body centerline from the nose to the c. g, distance parallel to the body centerline from the nose to the c. p.
pounds
inches
"Cg.
inches inches
'
'
lAX
X c~p.  X e.g. angle between the body :enterline and vI angle between the body centerline "andthe reentry flight path direction ballistic coefficient, W/CAS
y AY
flight path angle, negative for a reentry body change in flight path angle dispersion measured along the earth's surface refers to an increment except as noted angle defintd in Section 5, 4. 2  2731
S8
A 0 1
T14
Symbol go
,0
Definition
:c.
Subscripts!:i c E o R S. L, w refers to crosstrack component refers to conditions at reentry refers to conditions at zero roll rate except as noted refers to alongtrack component sea level refers to quantity resulting from wind effects refers to quantity resulting from density effect A dot over a symbol refers to a first derivative with respect to time. An arrow over a symbol means a vector quantity,
.1
274
'I
. .
.'.I': i
m .\
. . ..
'"
"
U
tHlE JOi~NNO NOPKINlS UNIVINSIlIt
spheric density profile, errors in mass, and aerodynamic asymmetries, The following examples are given to illustrate this discussion, Example I  Dispersion resulting from wind and deviation in density at low and moderate altitudes, Example 2  Dispersion resulting from wind and deviation in density at high altitude, Example 3

Dispersion resulting from a trim angle of attack for a missile rolling at a constant roll rate. Dispersion resulting from roll lockin at low altitude,
Example 4
Example 5  Dispersion resulting from a combined asymmetry for which spin through zero roll rate occurs, The major conclusions are: th Winds and departures from the assumed density are the chie; Nources of impact dispersion resulting from meteorological characteristics at the impact site. This dispersion increases rapidly as the flight path becomes shallow, as the missile velocity is decreased, and as the ballistic coefficient of the missile is decreased. The density effect is more sensitive to these missile and trajectory parameters than is the wind effect, Equations are presented for evaluating the dispersion for any wind and density profile,
I
:": . . . ..
!!
."
,, ....
. ..
. 7.
' ',
.... ..
""
.
'5
?"VM J M4
BtYtl~?
2. The equations derived for evaluating the dispersion resulting from deviations In density may also be used to evaluate the dispersion resulting from perturbations ih missile weight arnd drag, 3, The dispersion resulting from a vehicle that encounterm roll resonance (but remains intact) in likely to be small, On the other hand, the dispersion resulting that results asymmetry a likely combined havingIs vehicle a rate from in roll reversal to be extremely large.
27
I.
II
27.
"t.4'JOHNS
BILVI
SPUN.NMAEYTLAP4
Aiming instructions are required in order to deliver accurately a ballistic vehicle to a predetermined impact point. To obtain launch order s, a mathematical model containing all factors that influence the trajectory is constructed. However, even with the most elaborate trajectory simulation that can be devised, the model is not perfect and the vehicle departs from the intended trajectory, The magnitude of the miss distance from the intended impact point will be referred to as the impact dispersion. In Section 4. 1 some of the important factors which affect dispersion and which occur at the end of the powered phase were discussed (evror in terminal velocity and flight path angle). In this section we will discuss only those perturbations that occur during reentry. Accurate evaluation of dispersion requires use of a three or a sixdegreeoffreedom simulation. Simpler solutions, although lacking in accuracy, make it possible to isolate the important factors and provide insight into the general behavior of the various types of disturbances. These simpler solutions wfll be discussed here, 5. 1 Meteorological Characteristics The meteorological characteri'.ti.s that may produce dispersion are: 1. 2. 3. Errors in wind speed and direction, Errors in atmospheric density, Errors in atmospheric temperature,
Temperature is included since, in addition to its effect on density, it affects also the speed of sound and, thus, the aerodynamic coefficients that depend upon Mach number. However, the effect of the speed of sound is usually small (see Section 3. 1) and only the wind and density effects will be considered further.
t{
 277
.. . . . .. . ... ...
F.
5. 1.1
dispersion of uncontrolled reentry vehicles. A simplified analysis is given in Ref, 51, The important portions of this theory are summarized below. We make the following assumptions: 1. 2. The effect of gravity is negligible. The earth Is nonrotating and flat.
3,
4, 5. 6.
The wind is a horizontal head or tail wind (except as indicated), The reentry vehicle is aerodynamically stable (the center of pressure is aft of the center of gravity), The weight and CA are constant,
7,
A)
C~.HOR
IZO NT.AL
JE IS NEGATIVE
(27

278
S. .
....
... 
;
=:
 ,= =
me
Il
/i
In
gp,
rr
II j
til JOHNS HOPKINS UNIVM9IY CiLAIMnSP~iNQ MAMVIL~tND
I
The velocity vector is along the body centerline. The only force acting is the axial force, A, and it also acts along the body centerline (by definition), Suppose now the body encounters a tail wind. The body forces that act immediately are shown in sketch (b). The missile velocity
SA
NN
'V'.
awSHOWN , POSITIVE IS
(,
aii
7'W
1b)
vector and A are still along the missile centerline. However the velocity vector that produces aerodynamic forces This velocity vector on the missile is Vw"acts at an angle, / to hebody centerline and causes a normal force, N, which acts at the vehicle center of pressure. This force causes the body to pitch (weathercock) until the body centerline and VA/M are aligned. The conditions that exist when the weathercock motion has damped to negligible magnitude are shown in sketch (c).
I.2 1
279 
HORIZONTAL
4E
E
Vw
The normal force is now zero and A is along the body centerline. At all later times the body centerline remains aligned with / however, since A acts at a slight angle to I the missile acquires a drift velocity relative to the nowind trajectory. The force and velocity diagram including the drift velocity is shown in sketch (d). We will assume that the geometry shown in sketch (d) is representative of the reentry trajectory. At any given time (or altitude) the nissile velocity is assumed to be the vector sum of the nowind velocity for that time plus the drift velocity V R' Furthermore, it is assumed that VR <<V. Using these assumptions it is shown in Ref. 51 that tie dispersion resulting from winds is given by the equation: N
V~siny
13 I
VWFw
(51)
280
AMF A
aI
HORIZONTAL
v A/I
vi
vw
K
281
where
t iven in Trir:lc 5 1 l
ti sgivun in Tnbic 52, The index number~ j refers to a particular aititudu band. To account for a variable wind with altitude, the reentry altittude is divided Into N bands where the bandwidt', is small enough that Vw may be considered constont In each band,
1<
RANGE
The FA term Is a weighting factor that depends upon the altitude, altitude bandwidth, and KS., L. Since KS, 1_.
~sinV
L. an
~P
'Is
()
282
I
I t OHI I 0t00"III uNIVI'*1Y #4PLICrDPHYGICS StL1R SPIAOI~
LADORkTC'RY
MAA1LAPlh
a given altitude Fw for of atmos!!here, and type design and altitude bandwidth is a function of YE only. For most applications this variation may be neglected. When this is done the values of Fw should be evaluated at the lowest value of TEt used In the missile design. The dispersion Therefore, aldecreases rapidly with increasingdesi. though the percentage errore in 6 may be lrge when this procedure is followed, the magnitude of the error in 8 will be small. If the altitude bandwidth is constant, the magnitude of Fw it indicative of the sensitivity of 8w to winds in a given altitude barid. F'rom Tcble 52 it is observed that the most sensitive band corresponds to the 20 000 to 25 000 foot hand for NS, L.  1. 5, and the most sensitive altitude band decreases with increasingKs. L to the 10 000 to 1.5 000 foot band for KS. L.  6, 0. The calculation of the most sensitive band is treeated mathematically in Ref. 5I. Tne ballistic coefficient p varies with altitude and Mach number. Tne value of P at hypersonic speeds should be usrd for this calculation (also for 8 discussed in the next subsection), The summation term in Eq. (51) is called the ballistic wind. By definition, a ballistic wind is a constant wind (with altitude) such that the dispersion is the same as that resulting from a variable wind. If Vw is constant at all altitudes, then Eq, (51) becomes:
V H
,,
8M,
VE siin
1:3
(52)
The quantity p)..eceding the summation sign in Eq. (51) is called the range partial for Ainds. In Ref. 51, It is shown that Eq. (51) is valid for either head (or tail) winds or cross winds. If the wind
283
r..
S.I
is neither a head (or tail) wind nor a cross wind, the total dispersion may be obtained by computing the head wind and cross wind componerits of 6wand ad ding the results vectorially. Equation (5 1 ) has been found to be in excellent agreemetht with results obtanined using three or six degree of freedom trajectory simulations provided KS. L. is not gretter than about 6, For higher values of K8 L the equation overestimates 8 w. For applications where the body characteristics have been fully defined, the accuracy of Eq. (51) may be improved by the following procedure: 1. Using a three degree of freedom trajectory prograin, evaluate 8 w for, various values of VE and ^/E and a constant wind, Using these results and Eq. (51), 13 and plot it as a function of r. compute
2. 3.
Using the trajectory program, evaluate aw resulting from a constant wind in each altitude band. These computations are Made for one value of yE and the lowest IVtEI to be used for the missile. Using the results of step 3 and Eq. (51), compute Fw*
4,
These new values of 13 and Fw will differ slightly from those given in Tables 51 and 52. Using these new datai, the accuracy of Ow given by Eq. (51) may be improved,. . particularly for high values of Ks 5. 1. 2 Effect of a Deviation in Density The impact point of a reentry body is computed using some standard density profile (variation with altitude). If the actual profile is different from the standard
284
41
Therefore,
Using an analysis similar Lo that used for 8w, is shown (Ref. 51) that: 8,0 : 85}
it
M VE sinE
Y tan7
1 E N 5
&P5 F 0j
(53)
where 15 is given in Table 53 F is given in Table 54 44 is the deviation in density from the standard value (p) at a given altitude. All other terms are as defined in the previous subsection on the effect of wind. Again, F is a weighting factor that depends upon The most the altitude, altitude bandwidth, and KS. L. sensitive altitude band is the 25 000 to 30 000 foot band (see Ref. 51 for fuller discussion) and is nearly independent of KS, L.'
The summation term is called the ballistic density. By definition ballistic density is the constant value of ap/othat results in the same dispersion as the variable The quantity preceding th'a summation sign is called the range partial for density deviation. Equation (53) has been found to be in excellent agreement with results obtained using three and sixdegreeoffreedom simulations, provided KS L is not too high. In Ref. 51 it was shown that the condiftons rk.quired for the accuracy of Eq. (53) are those given in Figs. 51 and 52. When these conditions are exceeded, the dispersion given by Eq. (53) is too high.
IL
285
tS{_
For applications where the body characteristics have been fully defined, the accuracy of Eq. (53) may be improved following the same procedures given for winds in the preceding subsection. Since values of F. and Fw become very low at high altitude, one is tempted to conclude that the dispersion resulting from winds and deviations in density at altitudes above, say, 100 000 feet is negligible. For a high precision vehicle this conclusion Is not necessarily valid, since the magnitudes of Vw and A"/p at these altitudes are usually very large. It should be noted that, except for the direct effects (Ao/o and Vw), the variables contributing to dispersion resulting from winds and deviations in density are the same, However 80 is more sensitive to changes in the variables. 5. 2 Errors in Aerod ynamic Coefficients Errors or variation in aerodynamic coefficients res;ulting from reentry environmental effects (such as heating) may contribute to dispersion. The most sensitive coefficient from the dispersion standpoint is the axial force coefficient, CA.
I/ history during reentry,
In Section 5.1.2 we interpreted changes in K in terms of deviations in density. We could just as well have interpreted a variation in K as a variation in f3
Sicl
thenand
PAS
the__
II
and the Wc thenAK A Ws E dispersion characteristics computed in Section 5.1.2 are The equally valid for ,CA/CA,  AW/W, or  Ad/1, minus sign means that a given percent increase in W or Sresults in ae same dispersion as the same percentage decrease in pressure, density, or axial force coefficient. or sinYE. Since
286
I>,
I
THK JOHN$ HOPKINS UNIVSNSSITY
AK
287
.K
.,.
_ __ __ n m ~ n _p
m m
p  i u uv ~ n r i n
:1. , au
i n
5.4 Asymmetries
The problems associated with asymmetries were discussed in detail In Section 4.5. Asymmetries produce angles of attack; angles of attack result in lift and n change, in drag. For cases where ingles of attack develop and the missile is rolling, the major effect on 6 is the change in drag since the direction of the lift vector (in an inertial axis system) is constantly changing so that, over a period of time, the net effect of lift is "averaged out. " The effect of * for a rolling vehicle is usuallyjsmall unless resonance conditions occur. In this case o may become large and the vehicle drag will increase significantly, causing the missile to fall short of the intended target, This case can be treated as a change in CA so that the
A more
serious contributor to dispersion is the case where 0 and the roll rate becomes zero, Even for a very short period of zero roll rate, large dispersions may result from small angles of attack, Since, for this case, the lift vector remains fixed in space, the vehicle moves off course in the direction of the lift. We will consider these two cases separately. 5.4.1 Rolling Vehicle with Angle of Attack
If asymmetries result in a trim angle of attack and the missile roll rate is not zero, the major contributor to dispersion is the effect of the increased drag resulting from . The equation for the drag is: D =C qS (54)
CA cos'+
CN sin
 CA
CN sine
(56)
288
A'
THE
JOHNi
HOIPKINS UNIVtUITY
4 CD = C N
(57)
Therefore the effect of i may be estLmated using the procedures given in Section 5. 1, 2 where:
ICN
A A
(58)
W'When working with telemetry data, it is convenient to express ACD/CA in terms of body accelerations, Substi
(A()
&,
and there
"J
close
The dispersion resulting from tngle of attack for a nonrolling vehicle was considered in detail in Ref, 52 and the results are summarized below. Consider a vehicle with a fixed angle of attack that results in a lift force L, Consider an inertial axis system YZ as shown in the sketch. "Ifthe roll rate is low compared to the aerodynamic frequency, the lift (or normal) force is in the plane containing the asymmetric force (see Section 4. 5. 1). Therefore "L rolls with the body. As long as the roll rate is not to zero, the effect of L on dispersion is small since the direction of L (in the inertial axis system) Is constantly changing. However if the roll rate becomes
IL
 289
il
uNglVKOIr
Im I
ZY
.1
zero, even for an instant, the body will move off course in the direction of L. We wll consider in this subsection the case of a vehicle that is spinning at reentry but, under the influence of body asymmetries, the directior of roll rate is reversed during entry, The duration of "near zero" roll rate is very short. Assume that at b the roll rate passes through zero and that at this instant the lift is in the vertical plane. A change in flight path angle, A', results,
IORIZ.ONTAL H..
P'
19I NEGATIVE
a'?
tN
ISPOSITIVE
29
::
....
.,.
... , . ..
'
Thi JONNiI
MOP4iN3
uNIVll6ItY
.2......(51
'
in2 s in
0)
L Is in a plane SIf perpendicular to the vertical plane and if AV is interpreted as the resultant change in the direction of the velocity vector: h AV S.18
sinYl VE
(511)
SI
"
The impact pattern is elliptical with the major axis along the trajectory track and the minor axis across track. The ratio of the minor to major axis is I sin YEl'. The change in Y is given by the equation: Ay .dt (512)
g WV
COSY + V cosy V
Ldt
(513)
Equation (513) is valid provided the direction of L is fixed. However, since the direction of L changes over the time interval of interest, an effective value of .r Ldt must be used. Based on sixdegreeoffreedom simulation results, it may be shown that:
 291
li
2.11 6L.
Ldt 2
IrT I
Writing L
(514)
CL OoqoS
fLdt
(CN
 CA)
qS
(515)
Substituting Eqs. (515) and (513) into (510): 2,116hg (CN R = WV sin 2 E CA) I 0 q0S. (516)
CN
q S: (517
N (517)
WV
sin2
2
0 Io
+ gZ
2
0
N0
. 116hg 1 0(
( CA\
CN8
V0 sin ^141pof: Equation (518) is convenient for analysis of telemetry data since lo and go are obtained directly from the data and Vo and h 0 may be estimated using Eqs. (4. 322) and
0I
292
.0,.
I
YP[ J1M3HNIMOPKINS UNIVIURMSIT
M.RAm'tNB l@NG
(4. 35).
any combination of asymmetries, We consider next the dispersion that results from the particular types of mass and aerodynamic asymmetries (N and r) shown In the sketch.
0r
1:1
AXES) V (BODY
No
We will assume that the normal force resulting from r is very small so that the No results from an asymmetric moment defined by: no  m 0=C2+C2s "
The trim angle of attack, ao' is given b_ Eq. (227) where, in the notation of the present section, CITR a ao and Cm. Z!M .
0
qo, atocio, and 0  0o. Substi'uting these values where q into Eq. (516), writing po In teorms of H and Pop and noting that q 0 POIVO
293
S.
.....
.... .......'.
"TiL JOINII
NOPKINS
JNii.MSITV
APPL.IED
PHYSlC5 LADOOATC)PY
1.4 r
 C N /g w ,i ' 
SI X sP
1/2
0
(19)
Equation (4. 594) may be used to evaluate Ko and thus ho 0 o. Using Eq, (4. 594) and Eq, (227), 6 1 may be expressed in terms of io rather than Cm and r sin 0o, the result:
Sgiving
2'
A
6
1. 49i
1. 49oE 0
Po
sill Y
S." L. (520)
Equation (520) shows that for a given o (given Cm and 'AX/d primarily), the dispersion varies with h. (vartiations in r sin 00 if other variables are held constant) in accordance with the factor: oP K
0
S.
2 =f(h K (521)
P E h S. L. I e
PS. L.
At sea level, this factor is zero since h,) 0. The factor increases with increasing h. to a maximum and then decreases, becoming zerc at very high altitudes. The criti
type of atmospherc WS. L.) and upon KS.L. (or upon and/orYE). For the exponential atmosphere (Eq. (315))
294
it may be shown that E is a maximunr at ho Ho for very small values of KS L, and increases to ho = 2Ho for very For the 1962 Standard Atmosphre, large values of KS using 11 = constant' 3 000 feet, the critical altitude varies from about 23 000 feet at low KS.L. to about 46 000 feet at A plot of E versus ho is shown for several. high KS. L.. values of KS L. in Fig. 53. In summary, the following conclusiors were reached regarding the dispersion resulting from reversal of roll rate. 1. The dispersion may be in any direction. The of possible impact points is an ellipse centered at the locus nominal impact point. The major axis of the ellipse is along the flight path. The ratio of minor to major axis is
sin y.
2. The magnitude of the dispersion is a function of the many factors Listed below, making it impossible to draw generalized conclusions regarding the effect of specific variables. CA CAC CN m As Asymmetry
Noo
Aerodynamic
c.p.
W
X (Mass PE Initial Conditions
)
! Geometry
Type Atmosphere
S
S
I

295
i .i
 ,
,.
"
V.II i
.... f."
t.....
]l
UNIVKSIRITY
Only one type of combined asymmetries capable of resulting Ln & ind zero roll rate ir listed nabove. Hlowever, other combtined asymmetries such as C, and Cm o could also result in roll Late reversnl and Dirge dispersion. 3. For given nexmrovmyn:inic kI ielutcistics, mass, g.eometry, initial conditions, type of atmosphere, and Cmo, the Wo is fixed, However, h. (and No to a very small extent) varies with the values of r and 0o" The dispersion that results as r and/or 0 is changed is maximum when ho has some specific value. For the 19f12 Standard Atmosphere, the value of ho corresponding to 6 max is between 23 000 and 46 000 feet, depending upon the value of Ks. L. 5. 5 Errors in initial Reentry Angle of Attack and Body Rates
In some cases the initial reentry angle of attack, "i, and body rates, p, q, and r, are not known accurately, During the high altitude portion of the trajectory, these initial conditions affect the e history as discussed in Section 4. 4. If W, p, q, and r are not zero in the altitude r'gion where the aerodynamic restoring force is small., the missile experiences coning motion such that, in general, the body centerline does not rotate about the velocity vector. Therefore, in inertial axes, the lift force does not rotate about the velocity vector an it does for a rolling vehicle at altitudes where aerodynamic forces are predominant. During the early phase of reentry, initial values ofrs, p, q, and r cause dispersion resulting frorri both lift and drag effect,. The drag effect always tenois to cause the vehicle to fall short of the intended impact point. >lowever, the lift effect may cause dispersion in any direction depending upon the direction of The lift vector (determined by the location of the coning motion relative to the velocity vector), In general the locus of impact points resulting froi i those initial conditions is an ellipse whose major axis is approximately along the flight path. The ratio of the minor to major axis is approximately Isin Epl. The drag effect shifts the center of the ellipse toward the launch site from the nominal point (impact for nominal y, p,)q, and r).
I296
S...
3I~lIVN
SP*IN6
MARIL&ND
!
The vehicle motion is rather complex, and a sixdegreeoffreedom simulation is required to evaluate the magnitude of this source of dispersior', A typical order of magnitude of the semimajor axis for 6 resulting from ;; is 5 to 10 feet per degree error in 9. Therefore, the error is not negligiblc unless W is known fairly accurately. 5.6 Examples 5. G. 1 Example 1 Estimate as a function of E the dispersion resulting from a constant 100 ft/sec headwind (V 100) and a constant 10% increase in densLty (Ap/p 1.) from the 1962 Standard Density for the following conditions: V E2 20 000 ft/sec 1= 000, 200n, and 3000 lb/ft Since dispersion is most sensitive to the altitude region from 20 000 to 30 000 feet, a value of H 23 000 feet (see Table 31) is used. From Eq. (52): (100) x 000 w 23 000 sinE 3 w20 ,,I
15 sin3 y
From Eq. (53): 8=~ 6 123 000 20000 2 32.2x 0.1 x 15 2 sin V.Etan VE 4. 26 5 sin VEtan7E L. and_8
.
:1
From Table 31, P 2116 lb/ft 2 . Then K, 2116 6 For a given value of 0 and VE, 8, #sinsi~ yE
may be computed using 13 and 15 obtained from Tables 51 and 53. The results are shown in Fig. 54. This
 2072
297
b__
5.6,2
,xample 2
Assume that in targeting a mnissile no corrections than 1962 StandardDensity profile for altitudes greater from 100 000 100 000 feet. If the average wind at altitudes if the denand headwind ft/sec to 160 000 feet were a 300 profile, Standard 1962 the sity were 30% greater than of 3 the impact error resulting estimate as a function from these high altitude meteorological characteristics for VE x 20 000 ft/sec and YE From Eq. (51): 25n. Use H1 23 000 feet.
23 000(300)
4
.3 rj2WF
'I
44
F.
Ij~
iI
6~20 000X0
4
=155 15
422),
7765
34
r'
,! 2
E, Fw,
Using values of 13, Fw, 15, and F given Inr were computed ns a Tables 51 through 54, 6w and 8
w 0
298
SPO.NG
MARYLAND
function of KS L.T The values of corresponding to KS L. were computed using the equation: 2116 5000
KS. L.
KS. L.
Values of 6w and 6 are plotted in Fig. 55. The accuracy limit as given in Fg. 51 is also shown. Values of z: 1230 lb/ft 2 fall within the limits of accuracy; va.ues of 8 shown for p less than 1230 lb/ft 2 will be slightly high. This problem illustrates that bodies having low values of A tend to be sensitive to meteorological characteristics even at very high altitudes. 5. 6. 3 Example 3 For the reentry body havhLig the characteristics given in Table 21, estimate the dispersion resulting from a constant trim angle of attack ,if the missile has a con
istics are VE
stant roll rate during reentry. The trajectory characterUse H = 20 000 ft/sec and YE = 3023 000 feet.
For this case the dispersion results from the increase in drag, and the effect on dispersion is equivalent to a change in density. CN
(Eq. (58))
C 0' *o2
=A 2
0
0. 104 *
2 o
0
. 2 fore =19.2
0 = 0
in radians
0
in degrees.
,.i
S. L,
& F
0. 00585
2
Therefore:
=
200 23 000 2
)2
The missile will fall short bL9 feet for o= 1/20, 36 feet for &o = 10, and 145 feet for &o = 20. 5.6.4 E_.xample 4 Suppose in Example 3 the reentry body has an jo = 1 / 20 at reentry, the body passes through first resonance without lockin, but the asymmetries arj such that lockin occurs at an altitude of 10 000 feet and remains until impact. Determine the dispersion. Again, for this problem, dispersion results from the drag effect. However the angle of attack is amplified
by the resonance phenomena, From Eq. (4. 516), ,
S= Ai
1/D.
Using
0. 115
lr'W.
300
'I
Therefore: 8.7(1/2)

4._35
Substituting for K
P sin
P E PS. L.
KK
4.23
. .2,11
L..
At h = 2500 feet h = 7500 feet P/ P/P 0.0913 0.757 2. 21
S. L.
a = 2.420
From Exalple 3, the equivalent value of A&/0 is 0. 00585&o . Therefore, for h = 2500 feet, AL/p = 0. 0286; for h = 7500 feet, Ano/p = 0. 0343. Assuming that the value of 60/p at h  2500 feet is representative of the altitude band from sea level to 5000 feet and that the value of 6p/p at h = 7500 feet is representative of the 5000 to 10 000 foot band:
17
F j z1 6 Using Eq. 0 (53): 32.577) x 21 x0.00329
= 0. 0286(0. 049) + 0. 0343(0, 055) =
0. 00329.
20 000x0.5
20 feet
5.0.5
Example 5
The body having characteristics given in Table 21 ias the following asymmetries: 1. A center of gravity offset of 0. 042 inch.
S301
I I I I I "
2.
A body centerline distorted into an arc of a circle. The slope of the centerline at the body nose and base differ by the angle A&.
As a function of A&, assess the probability of encountering spin through zero roll rate, and estimate the dispersion that results if spin through zero roll rnte occurs. Neglect the possibility of encountering roll lockin at negative roll rate. Use the 1962 Standard Atmosphere with initial con30. ditions, PE I 1000*/sec, VE = 20 000 ft/sec and From Eq. (235), Cm (227): 57. 3 (0. 008A 8) =2&8 , 0. 0108 A8. From Eq.
0o where AG and
2(0,114)
O are
in degrees.
of K) at which the roll rate becomes zero is given by Eq. (4. 594). Substituting for the given terms: 1000 ; ~K/2 B (0.104)(0, 079) 20.001)( 3. 5 0.36 i
 3 sin 0 57.
9 sin 0~
4. 23, For
2116/1000 (0. 5) At sea level K : K zero roll rate at sea level: A sin 0
01
0.36  2,115
Since 0o may have any value from 00 to 3600 and assuming no correlation between the two types of asymmetries, the probability of encountering zero roll rate during reentry is:
3
U
302 
I
THR JOHN$ HOPKINS 61NIVCMSItY
I0
AO
8 sin
. 41
0
/,esin0
0.41 0. 50 0. 60 0. 70 0. so 1.0 2, 0
S. 0
0I Ao 0.O
0
The probability of encountering zero roll rate is shown as a function of Ae in Fig. 56. The altitude, ho, at which zero roll rate occurs is an important factor in the calculation of dispersion. For Ae ; 0. 41, ho may
be any value from sea level to some upper limit obtained when sin 00=1I or
in Fig.
roll rate occurs may be computed using Eq. (520). For a given Ae (or %O), the dispersion may be any value from 0 to some maximum value depending upon ho which, in
i..
"
ii
303
 3oW.
For low values of A6, the maximum vRlue of E (Eq. (521) and Fig. 53) correspond to the highest value of ho (00 = 900). However for the larger values of A6, the maximum value of E for this problem (KS. L.  4.23) occurs at hof 27 000 feet. Using ho data given in Fig. 57 and the E values given ;n Fig. 53, Emax for various values, of A8 are taoulatud below, h 0,e atE maX
0.41 0.5 0.6 0. 7 0.7 1.0 2.0 3.0 0 13 000 21 000 27 000 27 000 27 000 27 000 27 000 = 2 Ae: E max 0 8 600 10 800 11 300 11 300 11 300 11 300 11 300
AeE max
0 4 300 6 480 7 920 9 050 11 300 22 600 33 900
" 4(2AO)
max
m R maxe
1. 22A 9 E max .
j
"
The values of 8 Rmax may be computed using the listed values of Emax, and the results are shown in Fig. 58.
304
_.
________________,... ...
::::'?
. ,,,;,:.:
, .
I
I ?M JOgNSm HOPKINS UNIVIRthIY APPLIKD PHYSICS LABORATORY
IlLV911 ISNING MAPYLAND
The magnitude of the dispersiun is highly dependent upon 00. In the previous discussion we considered only the maximum effect. The probability of encountering a given magnitude for the major axis of the dispersion ellipse may be obtained by plotting 61 versus 00 for various values of A as shown in Fig. 59. The dispersions for values of 0o > 900 are not shown since OR is symmetrical about 00 i 900. The probability of encountering or exceeding a given value of ORfor a given A6 is obtained from the ratio A0o/360 where A0o is the range in values of 00 for which 8 R equals or exceeds the specified value. The probability plot is shown in Fig, 510, The values of P shown for 5R = 0 are the probabilities of encounterIng zero roll rate shown in Fig. 56. As an example of the interpretation of this figure, for A8  0. 50 there is a 19% probability that zero roll rate will be encountered; there is a 16% probability that 8 > 2500 feet; the maximum possible dispersion is 530Ofeet (the values of OR at P' =0 are the values shown in Fig. 58). Note that the dispersion that may be encountered if the missile spins through zero roll rate may be at least an order of magnitude greater than the dispersions considered in the previous problems, For example, the A& = 0. 50 asymmetry corresponds to a 10 trim angle of attack. It was shown in Example 3 that if the missilc does not encounter zero roll rate the resulting dispersion is about 36 feet for a. = 10. The spin through zero roll rate phenomenon is a serious problem to the designer of reentry vehicles that require good impact accuracy.
' .
...
305.
"
I
?rHe
JOHNS HOPKINI UNIVUlNsIY
REFERENCES
51
L. S. Glover, Approximate Equations for Evaluating the Impact Dispersicn Resulting from Reentry Winds and Deviations in Density, APL/JHU TG 1132, September 1970.
52
53
W. H, T, Loh, Dnamics and Thermodynamics of Planetary Entry, PrenticeHall Space Technology Series, 1963.
I.I
..:
I.
307

I[ :1 .;..
. . . ........ .,
,: ..!
!7
20 18 10 14
12
0
0 10
,
20
I
30
i
40
7'
I
50 (degrees) .60
I
70 80 90
Fig. 5.1
LIMITING VALUES OF K .L
309
TH
lSITYV
60
T r SPS.L,=
hi H
2116 Ib/tO
i
PS
STAN
40
02500
30
1000
20 2000 3000 10
00
10
12
14
16
18
20
22
24"!
310
'
"
!NK
JOHN$
k.
)PKING UNVlRSIyTy
20
oat
I hh
K *MaxA O
TEMPERATURE
K12SL,
4'
6
4
2
10
20
30
40
50
s0
70
80
ho (thouuandt of foot)
311
10 000
1000
1
sw (100 ft/imc
z 6p (10% INCREASE)
oo
100
"
I
2
__l
000.
I
30000
20
30
40
50
60
(dogrees)
70
80
.90
"7E
Fig. 54
312L/
4H
500I
400
(300 ftWas
HEADWIND)
8p 130%DENSITY INCREASE)
5+5)
J'30
.200
\\SA
\ ,
LIMIT FIG. 5.1 L ACCURACY
Iw
A
1800 2000 2400 2800 3200
80o
1200
~3(Ib/ft) I
Fig. 5.5 DISPERSION RESULTING FROM METEOROLOGICAL CHARACTERISTICS AT HIGH ALTITUDE
313
4k,
NN
I
jN
0 03
pa
314
IMIL'*
0
~IN
US
jiselospuno~i 04wwixv
IF
c
I
315
MANVLANIP UPOING
u:1
311
SS&.VUl
IC
000
dVd
7
1.1
MAM.AND
tt
0,15
0,:  3 .0 AO
0.3
a.
ol
0.2
0.50
V I.?
0.1
10
20
30 (thausarnds of foot)
40
50
Fig. 510
DISPERSION PROBABILITY
I.I.,
318
T49
JOHNS
14OPA0INO
UNIVIERIY
13
0 0.25 0.57 0.92 1. 33 1.84 2,44 3.20 4.13 5.25 6.60 9.90 :.
5.0
5.5 a8.17
6.0
319
Ii
II.
Altitude,1L
(thouands l . o ....
t.....
1 . 5 0, 007 n 2 3 4 5 (i 7 H 9 10 11 12 13 14 15 1, 17 0. 005 0.010 0.025 0,021 0, 030 0, 045 0, 0134
Valjuc,
nf IF
2. 0
0 0, 004 0, 002 0, 00 0,013 0. Ol1 0,017 0, 024 0. 033 0,053 0. 077 0,110 0. 071 0, O((i 0.105 0. 1.12 0. :10
, 0, 001 0, 002 0, 005f ol011 0. 0091 0. 015 0. 020 0. 029 0. 042 0. 066 0.100 0. 068 0. 085 0.114 0, 15B 0. 1.16 0.127
, 0 0, 00:3 0,001 0.00.I 0,010 0,0(t 0.013 0, 0161 0. 025 0. 0:4 6 0. 0113 0,0!17 0. 0U 13 0,0 1(G 0.117 0, 1[ill 0.1) 5 0. 1:12
0. 00Cl 0. 004 0. 009 0.023 0, 019 0, "27 0, 040 0, 058 , 071 0,.1 O 0,143 0. 08(1 0. 100 0.0. 90 0. 080 0,070 0,0 900
140I (10
S20140
100120 90100 8090 7080 ((70 60 4050 3040 25,3 0 20,5 1520 1015 510 S. 1. 5 Ntue.
O
0, 101 0(,147 0. 081O 0, 103 0. OB6 0. 0119 0, 0,8 0,047
ou0n
0. 11.1
T1 E'sL. data are applicable to a 1962 Standard Atmo)spher'e. Ilowuvr, th..v Wi v be used for auin, atmosphere by chnngillg the altltudc limits for ench hand. 'r' example, the .1 dantn npply to the altitude band from nn altitude corrusponid1mv to PP L . 0714 (610 000 feet for the 19102 Standard Atmompherc)to an altitude coript'e ncsPL ttO PIPs, L. 0.0443 (70 000 feet for the 1602 Stnndard Atmosphere).
320
_,I
TMS
JOHN*
UNIVURII0TY 8OPKINS
KSL
15
0 0.5
:
0 0.50 1.05
i. 85
1.0 t, .5 2. 0
2.5
3. 00
4.90
3.0 3.5
4.0
7. 55 11. 30
17 30
4.5
5, 0 5.5 6.0
27. 00
41.00 61.00 92.50
S321
"ti"
Table 54 Vanu#, of F A lttude (feet x 10 O ) 160140160 120140 100120 90100 8090 7080 6070 5060 4050 3040 2530 2025 1520 1015 510 05 p
KS . L.
_....
j
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1,5 0. 004 0,014 0.027 0.053 0.045 0. 055 0. 069 0. 088 0.118 0.142 0,140 0. 072 0. 068 0. 044 0.030 0.019 0.012
2.0 0. 012 0.014 0.023 0,046 0, 039 0. 050 0, 067 0, 085 0,111 0,138 0.146 0,076 0. 074 0, 048 0.032 0.022 0.017
:3,0 0. 009 . 00;) 0.017 0.035 0.028 0, 039 0. 058 0. 077 0.099 0.130 0.154 0. 083 0.082 0. 064 0.051 0.037 0,028
4.0 0,006 0.006 0.012 0.025 0.021 0. 029 0.045 0.065 0.090 0.123 0,157 0. 088 0. 090 0.078 0,066 0.053 0.046
5,0 0. 004 0.005 ,. 008 0.019 0,015 0. 025 0, 034 0. 056 0,087 0,120 0,158 0. 092 0. 098 0. 087 0.076 0.061 0,055
6.0 0. 004 0.004 0.007 0.017 0.013 0. 022 0. 030 0. 051 0. 086 0.118 0. 156 0. 095 0, 102 0. 091 0, 080 0.065 0,059
322
I
tHE JOHNS NOPI"IH INYiMRMITY
I I I I !
1
6. TRAJECTORY SIMULATION
I II
I SI
I
3231
I
i aa
SUMMARY
The discussion of the motion of ballistic missiles would be inconmplete wthout some comment on the "work horse" of all trajectory analysts  trajectory simulations. It is beyond the scope of this report to discuss any simulation in detail. However, for the uninitiated, a few brief comments are presented regarding the basic ingredients of all simulations, the relative operational cost of several types of simulations, and some characteristics of a sixdegreeoffreedom simulation currently in use at APL.
'.
Pp
a
325
1
... .. .. .
......
e.~
,I,
I
THE JOHN4 HOPKINS UINIVERSIT
A trajectory simulation is a computer program by which the time history of the vehicle motion may be determined. Given a set of initial conditions and the particular quantities for which a time history is desired (outputs), the reentry body simulation consists of five major parts (see sketch, page 328): 1. 2. 3. 4. 5. Equations of motion, Earth characteristics, Atmospheric characteristics, Mass characteristics, Aerodynamic characteristics.
Each of the major parts may be very simple or very complex. The equations given in Section 4, 3 provide the basis for a very simple simulation in which the equations of motion are reduced to a single degree of freedom (motion along a constant flight path angle), the earth is flat and nonrotating, the acceleration of gravity is zero, the atmosphere obeys the hydrostatic equation, the wind is zero, and the vehicle mass and aerodynamic characteristics are defined by a single parameter (0). However, the usefulness of this simulation is limited to those studies for which approximate time histories of a very few trajectory parameters are sufficient. For many types of work this simple simulation is not adequate. Increasing the complexity (and cost) over the onedegreeoffreedom (I DOF) problem, a threedegreeoffreedom (3DOF) trajectory simulation may be used. Usually the 3DOF designation refers to degrees of freedom in the three orthogonal linear dimensions. In this case the simulation is also called a "point mass" simulation. The only mass and aerodynamic characteristics used are the body weight and axial force characteristics. Any atmosphere and earth model may be used. This simulation is reasonably uncomplicated but has sufficient accuracy for most types of flight test analyses, for defining targeting data, and for evaluating most types of range partials (sensitivity of impact dispersion to various anomalies).

327
TH
LJ
LU
CC
u. U. L!J
0 . I. F
INI
cii
Il
32o
For evaluating the detailed dynamic behavior of missile motion and for calculating precise impact locations, a sixdegreeoffreedom (6DOF, three linear and three rotational) program is required. In this simulation, the body mass characteristics are defined by weight, center of gravity location (specified by three coordinates), moments of inertia, and products of inertia. The aerodynamic terms include the axial force, normal force, center of pressure location, pitch (or yaw) damping, roll damping, and asymmetric force and moment terms as required for a specific study. This type of simulation is the only one (of the ones discussed) that accurately accounts for angle of attack effects, Any atmosphere and earth model may be used. There are many modifications to these thrbe basic simulations. For a given study the analyst will select the type of simulation based on a tradeoff between accuracy and cost. For a typical reentry trajectory, the approximate machine cost of running the three simulations discussed above is as follows: lDOF 3DOF 6DOF Negligible, $1 per trajectory, $30 per trajectory.
As an illustration of a typical 6DOF reentry body simulation, one of the simulations currently in use at A1'L is described briefly. 6.1 Initial Conditions I. 2. 3. 4. 5. Geometric altitude, Missile velocity and flight path angle relative to rotating earth, and 0 components of (see Fig. 411), Initial roll orientation of the body axes with respect to the inertial axes, Body rates (roll, pitch, yaw). 329
.. .
. . .. . . . .
?He
JOHN$ HOPKINS
UNIVt61UIT
6.2
Weight, X, Y, Z coordinates of the center of gravity, Moments of inertia (about X, Y, Z axes), Products of inertia (about X, Y, Z axes).
2. 3. 4.
Provision is made for two sets of characteristics so that the trajectory for a twostage reentry body may be computed. Provision Is also made for varying the weight and the two lateral coordinates of the center of gravity as a function of altitude, This option is used for studies In which body shape changes as a result of aerodynamic heating. 6. 3 Aerodynamic Characteristics
* :
Some of the aerodynamic data are in an aerodynamic subroutine of the program and some are inputs, The subroutine provides the axial force coefficient, normal force coefficient, center of pressure location, pitch (or yaw) moment damping coefficient, and roll moment damping coefficient. The axial force is a function of Mach number, angle of attack, altitude, and velocity. The center of pressure, pitch damping, and normal force data are functions of Mach number and angle of attack. The roll damping coefficient is a function of velocity and altitude. Provision is made for using two sets of data corresponding to the two sets of mass data. The aerodynamic inputs include the asymmetric pitching, yawing, and rolling moment coefficients as a function of altitude and multiplying factors as n function of altitude for varying the aerodynamic data included in the aerodynamic subroutine.
330
I
?MK JOIINO ,4OPW0i, UNtVftSINIY
0. 4 Atmosphere CharacteristicB into the The 1962 Standard Atmosphere is built temperature, simulation but any variation of density, may be used. wind, and wind direction with altitude 6. 5 Earth Model model is The WOS/60 ellipsoidal rotating earth used, 6.6 Equations of .Moton
axes) used The six equations of motion (in body given in Elqs. freedom are the three rotational degrees of linear dethree the (4. 51), (4. 52), and (4. 53), and grees of freedom given in Eqs. (4, 44), (4, 45), and and moments are com(4. 46), The aerodynamic forces are then puted in the body axis system. Accelerations transand velocities computed in the inertial axes, and lations are computed by Integration.
6.7 Output
The outVarious output quantities are available. fcelowing the provides put format currently programmed time histories: 1. (body axes), Three components of acceleration Total angle of attack,
2. 3.
4. 5. 6, 7. B. 9.
APPLIrF
PHYSICS LABORATORY
The fiaal latitude and longitude are also listed. Several options are available regarding the printout time Interval. The integration time step is variable (with a minimum of 0. 0001 second), and readouts may be obtained at every time step, any multiple of the time step, every second, or any multiple of a second. For a 3DOF simulation only the first and second initial con ',,a, the first mass characteristic, the axial force coe at (as a function of Mach number and altitude only), and the three linear degrees of freedom are used, Hence the computer cost is lowered by an order of magnitude compared with the cost of using the 6DOF program,
'.,
J
i3
vi I.,
I,
...........................................................
IOV,2v
NOW,
MAmYAN0_
.'.
333
?He
JOHNS
HOPKINS
UNIVRPSIT
TypicaltUnits pounds
feet feet

b
iA
CA C N4 C
&
Cm
II,
q
D F g damping parameter (see Fig. 71 1) resultant transverse force resultant transverse acceleration,

pounds
2
g g g feet
longitudinal acceleration gy gz h * Iy K lateral acceleration normal acceleration altitude roll inertia pitch inertia parameter defined in Eq. (4. 36)
t2 slugft
slugft 2
3
Symbol m N P p r q or q S V W XI YoZ
T slugs
lUnits
pitch rate yaw rate dynamic pressure reference area velocity we Ight body coordinates with origin at center of gravity. X coordinate Is parallel to the body centerline, positive forward; Y coordinate is perpendicular to body centerline, positive to the right looking forward; Z coordinate? forms a righthand system with X and Y, positivo direction is down when looking forward, X X .distance cPpressure distance from nose to center of gravity from nose to center of
feet feet
angle of attack in XZ plane angle of yaw in XY plane ballistic coefficient, W/CAS total trim angle of attack
v
&X Ai
flight path angle X c.p. c~g. change in trim maneuver plane due to roll rate resonance index (see Fig. 711)

336
Symbol
Definition orientation of trim maneuver plane, positive measured clockwise from Y axis looking forward
Typical Units
degrees
WA
Subscripts
c, p.
center of pressure center of gravity conditions at beginning of reentry (400 000 feet) conditions when p  0 component in the X, Y, or Z direction, respectively
co g, E o X, Y, z
with respect to time. A dot over a symbol means the first derivative to time. respect Two dots mean the second derivative with
337
II
THIN IOINO HOPKINS 610NIVnsIty
SUJMMARY
A detailed discussion is presented ot the pocv(lures currently used at APL in the analysis of vehiclei.
The basic data available, Basic data processing, Conversion to engineering quantities, Special analyses for dispersion and "roll lockin.
Examples are given of: 1. 2, 3, 4, Typical flight test records, A method of extracting range time from these records, Typical body rate and acceleration traccs, Work sheets for computing angle of attack, aerodynamic pitch frequency, and several quantities related to roll resonance.
339'
7. 1 A vailiblo Dn2t:1
t1
:I In II
iI I f(;r, ti vrst wdo commonoly lv avlibe 1'iio 1q111ht J(I CFInaL ,vsis am( tleC p itch, yaw\, :11]L1 roll : tc .111 nl the
ctltvy dI spu e
is
d
i.It:i Iiflud1 (
Xf
.i t: 1111. tt 0
tho mass
T hes e include the wei;lit., center of grevity 10o!Iatlon (three coorI di ca te s , and tho ma rients aind r producets c nf Ui!ti1 T Ti!'
iriti;it eudIti.ons of. the roe nti'% tr,Jfjocorv ay c21 I120 also ru
viui(1
q.1 ir V !(.
400 000 feet ,ititude, which is an altitude wvell abovel thle beng~ibc atilosphere, and inclucle the velouity, flighit Path rate. Additional initial conditions must be, specified if the, refntry tb..'y motion at very high riltituatis 1,; being e~fxamined or 1.f the precise impact, point must bke deter. 1rrined. Thece quantities include the initial angle of attack, side slip aniglu, toll ang. , pitch rate, ancl yaw rate..1 Although in generai ail or thest, data tire required fox a precise ainalysis of' the reentr~y d,,rn:inics, it ocCP.sionalty happens that some part of the danta is nct availabLlec. When this situation arises, the anatysi s somietinmes m~ay be co.tinued without seriously compromising its accLacy by substituting nominal values for the mitssing .
data.
As a`iL npe the weather data may be unava.ilaI ble, but it may be possible to substitute weather data, (except for winds) that were, determined during prevituus tests in thait part of the world (and during that tim(! of the year) and still obtain satisfactory results, Or it may beI possible to ecoose one of the availabl~e standard atmospheres and obtain a. satisfactory result. Should thm acceleration at the reentry body center of gravity be unavailable, accelerations measured at other location. may be used to calculate the acceleration at the
.
341
center of gravity, An accelerometer pomitioncd nt n point away from the center of gra vity will bi influenced by the body angular rates and accelerntions; the data from such an accelerometer mA'y be corrected to yield thc necelern.. tion nt the center of grnvitY by nieans of the following
eq uations:
" measured g
mesrdg32X
Y measured g
3 z 2,
(pq
"
P) +
X
q 2+ r 22) (71)
2
Z 2",2327(rq + z
+

32.2
2 p+ )+" X
2
y 2 1 + 3(r !) q2)2
(+pq)
(72)
(p'
S"Zmeasure + q 2) 
+ qr)
2(pr
!)(7.:3)
7, 2
Data Sources
Redundant receiving stations are frequently used to ensure reception of telemetry from a reentry body. Ground stations, shipboard stations, and airborne stations may be used. The quality of the data received at any one of these stations may vary widely depending upon the geometry of the trajectory relative to the station location. As is now commonly known, a body reentering the atmosphere develops a plasma sheath that impedes RF transmission. The quality of the signal received through this sheath varies greatly with the aspect of the body relative to the station. Reentry bodies with a low ballistic coefficient (essentially a low weight to drag ratio) may have a plasma sheath of such impedance that transmission of the reentry data acquired during this "plasma blackout" must be delayed. It may be necc.;sary to examine the signal received by several stations in order to acquire the data needed for a reentry analysis. 7. 3 Data Format
Flight test data, as currently processed at APL, are in the form of either DITAR records or DITAP records. 342
Ilk.
In both cn ses the flight functions are digitized by dividing the calibrated range of the sensing instrument into 255 intervals. A value of the function may then be represented by a number of counts or bits ranging from 0 to 255. In the case of the DITAP records, the count level is plotted as a function of time, and the record is an analog display of the function, An example of the data format is shown in Fig. 71. Timing marks are applied to the DIITAP records according to the scheme In Fig. 72. In the case of the DITAR records, the bits are plotted In a condensed form that permits many telemetry functions to be displayed on one record. Two traces are required for each function, one trace to count from 0 to 15 and the second trace to count in increments of 16 from 0 to 240. By adding the counts from both traces, any number from 0 to 255 can be represented. An example of the data format for DITAR records is shown in Fig. 73. An example of the timing marks on the DITAR records is shown In Fig. 74. The series of marks on the margin of the record may be interpreted as time of day. The DITAR data format presents the data very compactly, and although a two point calibration is attached, it is sometimes difficult to establish the zero point from which one should begin counting the bits. From certain physical considerations, a value for one point on the flight re( ords can sometimes be established, and this point can be used to confirm that the correct zero bit level has been chosen, The following is an example of these techniques for confirming the. bit level. The pitch and yaw rate gyro data, before reentry, will have a sinusoidal shape. Except for the effect of principal axis misalignment, which is uuually small, the sine wave will be centered about zero rate. From the calibration tables used to convert to engineering units, the number of counts for zero rate can be determined, and this then fixes one point on the rate gyro data curve. Subsequent values can be determined simply by counting bits. Furthermore, the frequency of the pitch and yaw rate gyro sine wave is related
343
L:
"
~~iiLV~g
*PSNQ
MARYfLAND
to the roll rate by means of Eq. (4. 421) when the body is unaffected by the atmosphere. Therefore the exoatmospheric roll rate can be fixed by determining the frequency of the pitch rate gyro output, and subsequent values can be determined by counting. The count level for the accelerations may be established by noting that the exoatmospheric accelerations are zero. 'The bit level, therefore, must be the one that corresponds to zero acceleration. Subsequent bit levels can be estnblished by counting. In the case of the longitudinal acceleration data, advantage can also be taken of the two ranges of Instruments that are frequently used. One accelLi'ometer is sensitive enough to detect the low level of acceleration at the beginning of reentry, and the second is sturdy enough to measure the peak g. The less sensitive instrument may be expected to go off scale at some point in the reentry. The count level is 255 at the point where the range of the instrument was just exceeded. The acceleration at this point may be determined easily from the calibration and, of course, the less sensitive instrument should also be recording this level of acceleration so that, by means of the calibration, the count level for the less sensitive Instrument is established, Future plans call foi a computerized conversion.':1 of the data from counts to engineering units as well as for machine plotting of the data in engineering units. The flight data most commonly used for dynamics analysis are the roll rate and the three accelerations. Typical examples of these functions (as well as the pitch rate) are shown in Figs. 75 to 78. 7.4 Data Processing
0
Further data processing depends upon a knowledge of the flight Mach number history. The flight Mach number could be calculated based on the longitudinal dec'.leration if the altitude were known; however, the altitude history generally is not available. The recourse, then,
344
,i
:'
T"I JOINSt
WDOIKINl UNI""IvIt*TY MA O~ A AP D
simulation,
against the flight data to ensure an adequate simulation of the reentry trajectory, The easiest check that can be made is a comparison of the longitudinal acceleration history as predicted by the simulation with the flight data results. The velocity history of the body is obtained by integrating the acceleration history. If the simulated longitudinal acceleration history is verified by the flight data. and the transverse accelerations are small enough that the longitudinal acceleration represents the total acceleration, then the simulated velocity history (hence Mach number history) will be correct. For the purpose of matching the simulation results to the flight acceleration, it is convenient to compare the results based on the time after the occurrence of the peak g. The time from peak g to impact is a good check on the calculated Mach number history, If the acceleration from a nominal simulation does not agree with flight results, the factors in the flight that are not nominal (such as the aerodynamic asymmetries) must be included in the simulation in order to produce a match between the flight results and the simulation. Occasionally it is necessary to modify some of the aerodynamic coefficients in order to produce an acceptable match. At points where the altitude is known (generally the only point is at impact) the Mach number may be calculated based on the following equations. By definition of the axial force coefficient:
CAqS = A
or
(74)
C AS
(75)
.w
345 
.. .. .o
t!,I
tug JONN*
MOpKINS uNtIVURSTY
(7()oM) gx
is pri~ncipallyo a funotion
,I
of Mach number (and also a slowly varying function ofu anglt of attack and altitude). For a particular angle of attack and altitude, we can Plot CAdM2 versus Mach hsumbef eased on predicted aerodynamic coefficientsh We may then use flight data to calculate a value of CA M2 from Eq. (77), enter a. plot of CA M2 versus Mach number, and determine the Mach num ber. As mentioned previously, thixs seheme can usually be applied only at the impact point since the altitude (which determines P and hns some effect on CA) is not known at other points along the trajectory. With the Mach number history established by matching the simulation longitudinal acceleration to the flight results, the angle of attark may be computed as follows. The maneuver acceleration is related to the transverse force by Newton's second law: SF 2 2 W = whore g Fgz z + gy
(7s)
Expressing the transverse force in terms of the normal force coefficient, we have: CN rqS

(79)
346
I.
I
tHI JONN$ HOPKINS UNIVRlOrIY
jBfl
.1
CA' CA
'gx'
(710)
1
"gX CN The use of Eq, (711) with the flight values for the accelerations and the established (wind tunnel test, etc. values for the aerodynamic coefficients permits determination of trim angle of attack as a function of time, For routine flight test analysis, the envelopes of the normal and lateral accelerations are usually all that is extracted from the flight recorr's, and the mean value of the envelope at a particular point in time is used to calculate the trim angle of attack, A sample work sheet that shows a convenient technique for calculating trim angle of attack is shown in Fig. 79. An alternate technique for computing angle of attack is based on the rate gyro data. For small angles of attack: r + P (712)
q
o
qI
f+
(713)
(714)
For the times of interest, & and are frequently negligible so that ai and 0 may be obtained directly from rate gyro data. The significant advantage of this technique that neither the Mach number history nor the
347
S..
... . _ . . . . .  . . . . . . ... _.*. I ....... I I. i ." I ...
"
The trim i mneuver plane (i. u, the plnne of the transverse force acting on the budy) is simply calculated from:
1tan (715)
(4, 436): N WA ;
y rad/sec
(716) (
We may re
write this expression in terms of the flight data by substituting for qS from Eq, (710):
g ,,N &,x
A
CA
Iy
' w
(717)
Flight values of gX and the established values for the aerodynamic coefficients permit calculation of the aerodynamic froquency. A sample work sheet which describes a conv(,nient format for calculating WA is shown in Fig, 710, Depending upon the closeness of the aerodynamic frequency to the roll rate, the trim angle of attack may be amplified to a value different from the result that would be attained by a nonrolling body, as has been discussed in
348
I.
Section 4. 5. For some purpouses in reentry flight analysis, it is convenient to have a knowledge of the amplification factor. A sample work sheet that indicates the data necessary to calculate the amplification factor is shown in Fig, 711. The ratio of roll rate to aerodynamic frequency also affects the direction of the trim transverse force. A technique for calculating the change in direction of this force A6 is described in Fig. 712, 7. 5 Data Analysis A convenient technique for demonstrating the closeness of the reentry body to resonance is to plot both the roll rate and the aerodynamic frequency as a function of time on the same graph. As discussed tn Section 4, 5, when the roll rate is close to the aerodynamic frequency, the body is near resonance. Near resonance the previously mentioned amplification factors may be large, and large values of trim angle of attack may develop. It is possible that the large transverse forces associated with resonance (in combination with a center of pressure/center of gravity offset) may cause the roll rate to follow the aero>idynamic frequency. This effect, known as "roll lockin, " may be easily detected from the plot of roll rate and aerodynamic frequency versus time. Roll lockin is usually considered to be an undosirable phenomenon because of the large lateral loada, localized heating, and high roll rate associated with it (which may be sufficient to cause structural damage and which will cause some impact dispersion due to the abnormally high drag encountered). Of course, the roll rate may vary so that the rate reaches zero (a condition that usually leads to large impact dispersion), and the aforementioned plot of roll rate versus aerodynamic frequency displays this condition also. A te,'hnique for estimating the dispersion that results when a reentry body spins through zero roll rate .
346)
4k
?M
JONNS
HO*MIIII4gUNIvImsltv

sin 512 YO
V
"i' v,
feet fu
(71B)
rIP
If a simulntion for the basic tnajeetory is available, nd Vo aRV obtained for the trime when the flight
record indicates zero roll rate, (However, it usually is not necessary to simulate the roll or, transversc accelera
2.
Compute K o
P 0 Y /sin
using Eq. (4.322) aid the gx telemetry trace, 3, or, thus hIc 0f 4. Compute V0 using Eq. (4.35). VE CASY compute P and
350 
U
I
UNgvI~mUIV
REFERENCE
71
B. F. Fuess, "Impact Dispersion Due to Mass and Aerodynamic Asymmetries," J. Spacecraft and Rockets, Vol. 2, No. 10, October 1967, pp. 14021403.
351,I
"*
...
~~~~~~
m
In
IItUt,
It,,
66
I,, ""IIIQ
cc
z c D:z
m Du
SU.
"7
35
UNIVtINIIy
0Jw
La CC, l~
LL.
V) Ll.
~ S
2D
SIVOSRNGMOiR
Call
n41
m
tIo.
ml
Wri5
w
U.
'I
~tI
)
.. j~z
y
w
I
a,
U.
355
0on
0o j
01
r4
0
a2)
WW'N MV
8z
1Lu w
2
<
U
I'ec
z.

..
U j
'U
36
I
?NI J0H.I
APPLIED
PiYSICS
IPU..G
I,vum
'I
11
F w J
a
AZ
r4 U
(pUOKh3IS.h511.LVII
1iOh
I
II
............
wj
I I/I
I'J
0
cc
Ii
ID
sz
L
16)N~lVU3300
58
,
uJI
9
wh
!I.
(5) N0I.LVW313D0ViviuflINDO1
359
S~vlmSPURS
M~~hAA
44
0010
t
WAS NOTACCURAIE
GO
z
u
'
"2
G u
'QB'rAINtD
z
(DIMENSION LESS)"']
FROM COMPUTER SIMULATION. USE TIME OF PEAK LONGITUDINAL DECELERATION TO MATCH FLIGHT DATA TO SIMULATION (DIMENSIONLESS)
Q00100
LONGITUDINAL DECELERATION (g)
01 3 !
,,U)
Fig. 79
361 
.i
...... .. ... . .. . . . . . . . , . . . , ,
@x@ W'A
@' K
K'.
3 J/,,W
01
8 x(
Xc~g'

CENTER OF GRAVITY OBTAINED FROM PREFLIGHT DATA (INCHES) OF PRESSURE OBTAINED FROM LISTING OF AERODYNAMIC
SCENTER
p.
CHARACTERISTICS (INCHES)
(D xG
_____
AXIAL FORCE COEFFICIENT OBTAINED FROM LISTING OF AERODYNAMIC CHARACTERISTICS (DIMENSIONLESS) OF NORMAL FORCE COEFFICIENT SLOPE OBTAINED FROM LISTING AERODYNAMIC CHARACTERISTICS (PER DEGREE) FLIGHT TEST, LONGITUDINAL DECELERATION (gI) OBTAINED FROM COMPUTER SIMULATION. USE TIME OF PEAK LONGITUDINAL DECELERATION TO MATCH FLIGHT DATA 1O SIMULATION (DIMENSIONLESS)
SCA
WA
57.3.A__Cl
(XI.
h AX)x
Fig.7.10
362
T"N%
UILYS,. I~iN
AMPLIFICATION FACTOR
14
CMCI
PCOBTAINED
PIEVIOUFROM
OALCULATIO
CH@CTMSTC IDMESINLSS
CN t R LOPE OBTAINED FROM LISTING OF AERODYNAMIC T OBTMAL FORCE COEFFICIENT CHARACTERIOMTICO PERRAILAN)
x N"JMU. A SC
II
PITC DAPOLINGCEFFIACAD9IPITC FROM IBANED RTMISTN FAROlAI D) DATA ( FLIGHT ARCTION DECELE LONGITUDINAL FROM LIS'TING' O'F 'AERODYNA M'IC NDINION~iBTAINED FORCE ITAXIAL OE r F ICI'EN CH RAT
l
F@AF.RODYNAMIC PRQUNC
" 4
2x Iy (PITCH INERTIA)m Ix (POLL INERTIA! AND OBTAINED FROM PREFLIGHT DATA (degre/econd)
FLIGHT DATA (degreeCsiscond) ROLL HATE VELOCITY OBTAINED F'ROM COMPUTER SIMULATION (fwlsa/conll) F~REQECJ C, NUMBE.R TIME BEFORE IMPACT,,
SMACH
TIME OF PEAK LONGITUDINAL COMPUTERLSIMULATION, FROM TO' ObTAINED DECELERATION MATCH FLIGHT DATA TO SIMULATION IDIMSNSIONLESS)
__
. I C(.
t,
A
p2  2 I1.y)
I D.
..
F ig. 7.11
.,
 363 .
Vi.
7..i
AlP
!I
SIN A
A W 0.
0CL 0bi
61)
03 O TI ME
IMPACT
(COL. GJ FIG.7,11T
(CUL1
3 1N Al
Fig.712
MA
"
,laj,.

. .
364
two aw
SS.
"1
ii.
MIL
"U

205
,I ,
, ,
?TH4 JOHN$
HPKINS
UNIVSINYV
Conversion Equations
Degrees Centigrade 5 ( degrees Fahrenheit

32)
1, 94 slugs/it3
3 1.94 x 103 slugs/ftt
2, 09 lb/ft2
1 knot
1, 69 ft/sec
36
L
2
g :32. 17 ft/sec
2
Constants
Acceleration of gravity at sea levl Ratio of specific heats 1. 4 ft2 sec 2 degrees
j.
degrees
Rankine
Radius of earth = R = 3441 nml we 2.09 x 107 feet 0. 729 x 104 rad/sec
4
2. 718
i.
Itt
.I
V!
368.
I
i
II
M "osUN T
VQIWS MAftT6AND
INDEiX
36
JOH 911d
H PLIM6t ~NS
URSYIATV
~SILVIR
Accele'ration, Lateral
a convergence

rolling
152 154
Aerodynamic Coefficients Asymmetric moment for flap on a cone Asymmetric moment for a. distorted cone Definitions Equations for a cone at high Mach numbers
Aerodynamic Forces
A symmetric Axial Drag Lift Normal Aerodynamic Moments Pitch Pitch damping Roll damping Aerodynamic Pitch Frequency Altitude Geometric Geopotential Amplification Factor Angle of attack Roll rtcceleration
371.i
171 148
1 51
1 3, 155 26 81 86
!
Center of gravity or center of pressure off set EWect on roll rate In plane Moment (couple) Moments of inertia, unequal Out of plane Produ,ts of inertia Types Atmosphere Exponential Polar Standard Tropical Standard 1962 Standard Axial Force Azimuth, Inertial B Ballistic Coefficient Ballistic Density Ballistic Wind Barosphere 572117 285 283 49 4 174 180 177 168 217 1*77 207 167 A 55 66 66 68, 69 13 103
tH I
! 1APPLIED
JOHNS
14OPKING
UNIVEEU2TV
PHYSICS L 'BORATORY
Boa5, Rate Effect of product of inertia in vacuum Transverse Effect of unequal pitch and yaw moment of inertia Boundary Layer C Center of Pressure
13
Cone
Aerodynamic coefficients Space Constants and Conversion Equations Continuum Flow Coriolis Force D Damping Pitch Roll Deceleration, Reentry History Density Ballistic Polar Standard Atmosphere Range partial for Tropical Standard Atmosphere Weighting factor for 1962 Standard Atmosphere 285 66 285 66 2B5 68, 69 23, 24, 29 140 367 20, 260 59
'
373
1' ~..
 . . . . . . . . . . . . . . . .. . '. " :'.. . . . .,,: *=_
. 
THi
Dispersion Axial force coefficient errors Density deviation Missile weight error Trim angle of attack  rolling missile  nonrolling missile Wind Drag Dynamic Pressure Definition Reentry history Dynamic Stability E Energy Equation, Conservation of Equations of Motion Sixdegree offreedom  symmetrical missile  asymmetrical missile Exosphere
17
119 146
81
137 168 49
Flight Path Angle Flight Path Angle, Inertial Flight Test Analysis Flight Time Ballistic trajectory in vacuum Reentry history
81 103 341
84 116
FreeMolecular Flow
20, 260
374
;L
.. .
.............
?He
Frequency Aerodynamic or weathercock Body pitch  in vacuum G Gravity Effect on ballisti.c trajectory Law of Variation with altitude H Heat Absorbed at stagnation point Rate at stagnation point Homosphere Humidity, Effect or. Density Hydrostatic Equilibrium Equation
I
147 139
81 81 68
Inertia
Moment of Product of
"KnudsenNumber
L Lateral Acceleration
19
152, 154
375 
Lock In
Condttlons required for Definition Effect of roll damping Effect of roll torque 188 186 198 198
119, 120
17
68 65 65 83 56, 67
Mean Free Path, 1962 Standard Atmosphere Mesopause Mesosphere Minimum Energy Trajectory Moisture, Effect on Density Moment Asymmetric Pitch Pitch damping Roll damping Moment of Inertia, Unequal Momentum, Conservation of Angular N Normal Force
28, 29 25 15 16 21 7 81
13
52
376
!V.
?MR JuINIG
HOPKINS UNIVgRIgTlt?
Pitch Frequency, Aerodynamic Polar Standard Day Pressure, Dynamic Pressure Force, Horizontal Pressure, 1962 Standard Day Product of Inertia Criterion for lock in
147 66 17 59 68
215
Definition
Effect on body rate history, vacuum Effect on trim angle of attack
R
207
208 211
55
Range Partial
Density
285
Wind Range Sensitivity Factors Reentry Dynamic pressure Heat absorbed Heating rate Longitudinal deceleration Reynolds number Time Velocity Resonance, RollYaw
283 87, 93, 94 119, 119 119, 119, 119, 118 118 120 120 120 120
Definition
Effect on Effect on Effect on First and roll acceleration trim angle of attack windward meridian second
171
180 171 173 187
I'f
Sa377S.
.. . .. . .. .. . . .. . . . . .
T?4
18 119, 120 56 16
Effect of aerodynamic asymmetry Effect of centerofgravity offset Effect of product of Inertia Effect of resonance Effect of unequal moments of inertia Equilibrlum
Scale Height
Definition
1962 Standard Atmosphere Second Resonance Simulation Slip Flow Space Cone Speed of Sound Definition Polar Atmosphere Tropical Atmosphere 1962 Standard Atmosphere Spin Up, Spin Down Standard Day Polar Tropical 1962
55
68 187 327 11, 140 6I1 66 66 68 187 260
66 6"68, 69
378I '
tfm
14 14 65
65 35
T
Temperature
Polttr Standard
Standard Tropical 1962 Standard
66
66 6B, 69
I"
I,
A'1
r379
Viscosity Definition
Sutherland's equation 19[62 Standard Atmosphere W
18
35 68
Wind
Ballititc
283
60 283 282
380 